You’d have to live in a cave to avoid the push toward artificial intelligence in recent years. The future of business is on the cusp of a dramatic transformation, and AI is at the forefront of this revolution. This cutting-edge technology has the power to reshape industries, disrupt markets, and redefine the way we do business.

As we embark on this incredible journey, organizations need to understand the potential impact of AI on their operations and embrace the opportunities it presents. Although AI technology has been around for decades, it has only recently become more affordable and capable. However, the downsides of AI include higher computing costs and misalignment with business needs.

What Are the Benefits of AI?

AI can revolutionize business processes and workflows by automating repetitive tasks and augmenting human decision-making. This means increased efficiency, accuracy, and productivity across various business processes, which can and should translate into fewer humans on the payroll.

Imagine a world where customer service is handled by AI chatbots that can understand and resolve complex queries from customers. Or, where data analysis is performed in real-time, which enables businesses to make data-driven decisions faster than ever before. It’s no surprise that AI has the potential to unlock immense value and streamline operations in ways we are just beginning to understand.

AI’s fundamental purpose is to enhance our understanding of data and unlock valuable insights from massive datasets. Organizations can use machine learning and predictive analytics to uncover hidden patterns and make proactive decisions. This leads to a much better understanding of your business and, thus, the ability to improve it with almost 100 percent foresight, which just a few years ago was an educated guess.

This visibility allows business leaders to anticipate customer needs, optimize supply chains, and identify new market opportunities. The availability of this information can change them from fast followers to innovators and even disruptors. A better product and customer experience usually leads to more revenue and the ability to dominate the market.

Also see: Top Generative AI Apps and Tools

What About the Downsides of AI?

While artificial intelligence holds immense business potential, it is important to acknowledge its downsides and potential challenges. Those who are implementing this technology are starting to see a few key areas of concern:

Increased Costs

Implementing AI technologies can be costly, requiring significant investments in infrastructure, training, and expertise.

Those who advocate for AI often cite cloud computing as a cost-effective technology implementation method. That is not the case.

Hardware costs have decreased significantly in the last few years, which is driving many businesses away from public cloud providers to run AI systems in traditional enterprise data centers where the operating costs seem to be lower.

Regardless, businesses that lack in-house expertise will need help with the financial burden of adopting and integrating AI into their operations, especially small- and medium-sized businesses.

Data privacy and security

AI systems rely on vast amounts of data. The information needed to train algorithms is often of a sensitive and personal nature.

It is critical for enterprises to ensure data privacy and protection to avoid legal issues that may arise if they fail to monitor their data usage…or even the data that these AI systems generate.

Businesses need to understand that they must establish robust security measures and comply with applicable regulations.

Ethical concerns

AI algorithms are only as unbiased as the data they are trained on. Without careful attention, biases and discrimination can emerge that will impact decision-making processes.

Indeed, there are court cases now that focus on just this problem, where AI decisions are biased toward people of certain races and ages. Some businesses lost millions in court settlements due to proven bias in their AI systems.

Checking for fairness could have prevented this. Striking the right balance between automation and human oversight is crucial to ensure the ethical use of AI technology.

Also see: Generative AI Companies: Top 12 Leaders

Understanding the “True Value” of AI

The impact of AI extends beyond operational improvements. It has the power to transform customer experiences completely. AI-powered personalization means tailoring products, services, and interactions to individual customer preferences. Those of you involved with digital transformation projects will recognize this as the end goal of digital transformation.

Using personalized recommendations, targeted advertisements, and intelligent customer support, AI has the potential to create highly customized and engaging experiences that drive customer loyalty and satisfaction. We saw this in the early days of using AI systems for recommendation engines in commerce platforms. These could double sales in many instances and are part of most selling websites today.

Before using AI technology, it’s important to weigh its tradeoffs, costs, and risks. That’s the case no matter what new technology you employ, but there could be a significant amount of good and evil that you can do with this technology in a very short period. That makes AI more variable in terms of what value or damage will or could be returned to the business. Over the next few years, we’ll understand more about AI’s emerging best practices.

As with any disruptive technology, embracing AI comes with challenges. Organizations must address concerns about data privacy, ethics, and the impact on the workforce. They must invest in robust data governance and security frameworks, as well as ensure the transparency and accountability in AI systems, and re-skill or up-skill employees to work alongside AI technologies. Change management becomes crucial to successfully integrate AI into the organization’s fabric.

The rise of AI brings immense potential and transformation to businesses across industries. However, we need to consider the good with the bad. Embracing the disruptive power of AI can unlock new levels of efficiency, optimize decision-making, and deliver highly personalized customer experiences.

However, organizations must approach this technology strategically, addressing challenges head-on and investing in the necessary resources for success. I would lie if I told you that AI had only an upside.

lso see: 100+ Top AI Companies

The post How AI Will Change and Not Change Your Business appeared first on eWEEK.

On Tuesday, October 17 at 11 AM PT, @eWEEKNews will host its monthly #eWEEKChat. The topic will be the future of cloud computing, and it will be moderated by James Maguire, eWEEK’s Editor-in-Chief.

We’ll discuss – using Twitter, now called X – current and evolving trends that are shaping the future of cloud computing, a dynamic platform that is now foundational for the enterprise. Our ultimate goal: to offer guidance to companies that enables them to maintain a competitive strategy as cloud continues its rapid evolution.

See below for:

• Participant list for this month’s eWeek Tweetchat on data analytics
• Questions we’ll discuss in this month’s eWeek Tweetchat
• How to Participate in the Tweetchat
• Tentative Schedule: Upcoming eWeek Tweetchats

Participants List: the Future of Cloud Computing

The list of experts for this month’s Tweetchat currently includes the following – please check back for additional expert guests:

• Trent Fierro, Cloud and AIOps Solutions Marketing, Aruba 
• Andi Mann, Global CTO and Founder, Sageable
• Zeus Kerravala, Founder and Principal Analyst, ZK Research
• James Maguire, Editor-in-Chief, eWeek [moderator]

Tweetchat Questions: the Future of Cloud Computing

The questions we’ll tweet about will include the following – check back for more/revised questions:

1. I’m just checking: in 2023, the only tech topic that people discuss is AI. So…yawn…does cloud still matter? (Hint: yes, more than ever.)
2. What key trends are driving the cloud sector here in late 2023?
3. What’s the most vexing cloud computing challenge today? Managing multicloud? Containing costs?
4. How do you recommend companies address this difficult challenge?
5. Okay, to the future: what shifts do you see in cloud as it evolves over the next few years?
6. How can companies best keep up with – or stay ahead of – these changes in cloud?
7. How will cloud computing be influenced by AI and generative AI?
8. What about the future of cloud and a related technology (besides AI)? How about the future cloud and edge, or cloud and data?
9. Your longer term sense of the cloud? What’s off in the misty distance?
10. A last Big Thought about cloud computing – what else should managers/buyers/providers know about enterprise cloud?

How to Participate in the Tweetchat

The chat begins promptly at 11 AM PT on October 17. To participate:

1. Open Twitter in your browser. You’ll use this browser to Tweet your replies to the moderator’s questions.

2. Open Twitter in a second browser. On the menu to the left, click on Explore. In the search box at the top, type in #eweekchat. This will open a column that displays all the questions and all the panelists’ replies.

Remember: you must manually include the hashtag #eweekchat for your replies to be seen by that day’s tweetchat panel of experts.

That’s it — you’re ready to go. Be ready at 11 AM PT on September 12 to participate in the tweetchat.

NOTE: There is sometimes a few seconds of delay between when you tweet and when your tweet shows up in the #eWeekchat column.

#eWEEKchat Tentative Schedule for 2023*

July 25: Optimizing Generative AI: Guide for Companies
August 15: Next Generation Data Analytics
September 12: AI in the Enterprise
October 17: Future of Cloud Computing
November 14: Edge Computing Trends
December 12: Tech in 2024: Predictions and Wild Guesses

*all topics subjects to change

The post eWEEK TweetChat, October 17: the Future of Cloud Computing appeared first on eWEEK.

AIOps, which stands for artificial intelligence for IT operations, is a growing industry practice in which IT professionals use AI, machine learning, and automation techniques to improve their workflows, with a goal of improved efficiency and standardization.

Businesses everywhere are adopting the practice because, in theory, it should make their jobs easier. However, AIOps is a complex process that, when executed without a clear plan in place, can lead to more inefficient and convoluted IT processes.

To achieve the best results when first implementing AIOps in your organization, it’s important to understand not only how AIOps can benefit your team but what AIOps best practices you should follow from the outset.

AIOps Best Practices: Table of Contents

1. Invest in an AIOps Platform That Integrates With Your Existing Tool Stack

2. Define Concrete AIOps Operations and Goals in Advance

3. Identify Relevant Data and Data Sources

4. Maintain High Data Quality Standards

5. Don’t Skimp on Data Security Best Practices

6. Complete an AIOps Test Project

7. Establish an AI Use Policy for Your Organization

8. Regularly Monitor Your Network and AIOps Workflows

9. Document AIOps Processes as They Are Established

10. Don’t Limit AIOps Best Practices to IT Use Cases

Bottom Line: Implementing AIOps Best Practices

The Top AIOps Best Practices

1. Invest in an AIOps Platform That Integrates With Your Existing Tool Stack

AIOps is all about making your current artificial intelligence and IT processes more efficient, and that only happens if your AIOps tool(s) integrate with all of the other most important resources, data sources, and business applications in your tool stack.

For the best possible results, look for AIOps platforms with deep native integration libraries or a capable API that makes it easy to connect your AIOps platform across your IT ecosystem. Aim for minimal customization and fine-tuning requirements, at least at the start, so you can keep the AIOps platform running efficiently even as your other tools change and scale.

2. Define Concrete AIOps Operations and Goals in Advance

AIOps is only effective if you know what you’re using it for and what you’re hoping to improve with AI-driven automation and workflows.

Every organization’s goals and KPIs are different when it comes to AIOps, but regardless of how the individual goals look, you’ll achieve better outcomes if your goals are framed as part of larger IT, operations, and business objectives.

For business leaders who are not sure where to start, consider setting goals around the following IT operational tasks and improvements:

• Improved mean time to resolution (MTTR)
• Improved mean time between failures (MTBF)
• Reduction of IT incidents and outages, particularly for critical services
• Resource optimization and allocation across cloud and on-premises resources
• Proactive issue detection and incident mitigation
• Reduction of false positives and false negatives
• Resource provisioning time
• Capacity utilization across usage metrics
• Quality and frequency of change impact predictions
• Number of policy and compliance violations over time
• IT operational expenses over time

3. Identify Relevant Data and Data Sources

Once you’ve completed initial tool research and goal setting, you’ll need to get a handle on all relevant data sources that need to be accessed, integrated, and migrated for effective AIOps.

Consider the different formats of data involved, how this data is used, and the different location(s) where this data currently lives in your network. Most important, determine if this data exists in different formats across your organization; if that’s the case, you’ll want to clean and standardize that data before adding it to your AIOps system.

4. Maintain High Data Quality Standards

AIOps relies heavily on data for both AI/ML training and operational queries. It’s important to hold your organization to high data quality standards so all data that you need is accessible, understandable, and in an actionable format.

Some of the steps you can take to improve data quality include using data cleansing, data mapping, data preparation, and data integrity management tools. You can and should also invest in data governance tools, as these tools can help improve data quality, security, and compliance while also assisting IT teams in monitoring and addressing changes in data quality over time.

5. Don’t Skimp on Data Security Best Practices

Data security best practices and tools are necessary for smooth operations, regulatory compliance, and maintaining business reputation and customer trust. Beyond investing in data governance tools as mentioned above, your AIOps teams should adhere to the following data security best practices as well:

• Use only secure and trusted databases and data sources.
• Keep your data storage systems updated and make sure all data storage practices align with regulatory and security requirements.
• Set up mobile device use and data management policies for employees.
• Designate a multi-tier access and authorization system for your most critical assets.
• Whenever it’s necessary for larger groups of people to access or use sensitive data, look for additional ways to protect that data, such as data encryption, anonymization, and/or masking.

Bear in mind that if your organization is interested in using generative AI in its AIOps workflows, additional data security rules and best practices may be necessary.

Also read: Generative AI and Data Analytics: Best Practices

6. Complete an AIOps Test Project

Before setting up the entire network and several automations within an overarching AIOps framework, you should first complete a test run with a smaller project as a proof of concept. During this test run, you can monitor how the infrastructure performs and also make note of how the team handles this new workflow. From there, you can make adjustments to infrastructure, team training, cybersecurity tooling, and the AIOps plan accordingly after the test project is over.

7. Establish an AI Use Policy for Your Organization

Any organization, department, or team that uses AI needs to be held to an AI use policy and receive training on the technology and its role in operations.

The policy should particularly address questions about how AI is being used in the organization, individual roles and responsibilities when using AI, and how to maintain data security and integrity while using AI. If you’re looking for an example AI policy to use as a foundation for your own, this generic artificial intelligence ethics policy may be a good place to start.

8. Regularly Monitor Your Network and AIOps Workflows

Your team should be monitoring network performance at all times, but it’s especially crucial when implementing a new operational practice like AIOps.

On a regular basis, you’ll want to use monitoring tools with features that support analytics and deeper dives into AI and ML performance. On a less frequent basis, your team or an objective third party will need to conduct deeper network audits to assess how each piece of the AIOps workflow is performing and how or if it’s impacting the rest of your network’s performance.

Some monitoring tools simply notify users when a possible issue arises, while others offer suggestions or take corrective action. You’ll want to select a monitoring tool that complements your team’s skill sets, budget, and other team requirements.

9. Document AIOps Processes as They Are Established

Documenting AIOps processes after they are quality-tested and approved is one of the surest ways to protect the integrity of the overall operation.

Concrete, detailed documentation also supports change management efforts, giving business leaders the resources they need to keep things running smoothly as team members, tools, and other components of the workflow change over time. Documentation can be kept in traditional data storage and database systems for easy access, but many businesses will benefit from storing this documentation in an additional, off-network or highly secure location.

10. Don’t Limit AIOps Best Practices to IT Use Cases

AIOps is technically focused on IT operations, but these best practices can be applied across an organization, especially as AI and advanced business technologies become a normal part of enterprise operations outside of IT departments.

Making sure all members of your organization across all departments understand how AI can automate their workflows and create new efficiencies could help your overall organization get roles and responsibilities, tool sprawl, and inefficient daily operations under tighter control.

Learn more about why AIOps best practices should be implemented across different business departments.

Bottom Line: Implementing AIOps Best Practices

AIOps is still a fairly new IT operations strategy, but because of the automations and new efficiencies it introduces to team workflows, it has quickly gained traction across industries and teams.

Despite its advantages, IT professionals need to remember that setting up AIOps can be a complex, multi-faceted process and getting it wrong may create even bigger operational headaches than existed before, not to mention potential new security and compliance issues.

If your team takes the extra time necessary to properly set up automated workflows with appropriate expectations and rules from the beginning, your AIOps plan stands a better chance for success in the long run.

Read next: Best Artificial Intelligence Software

The post 10 AIOps Best Practices appeared first on eWEEK.

I spoke with Krishna Subramanian, President of Komprise, about new findings from a survey of IT professionals about unstructured data management; the survey responses provide key insight into issues around data and AI.

Among the topics we discussed:

• The report notes that “As the generative AI marketplace expands and executives push for departments to leverage new solutions for competitive advantage, the need for an unstructured data governance agenda is strong.” Why is this true?
• Additionally, “To cope, enterprises are restricting the AI tools and/or data that employees are allowed to use, according to the survey. IT leaders are also pursuing a multi-pronged approach for mitigating risks of unstructured data in AI, encompassing storage, data management and security tools as well as internal task forces.”

Highlights of the Survey:

• Most organizations (90%) allow employee use of generative AI yet 66% of organizations cited top data governance concerns of preventing security and privacy violations, lack of data source transparency leading to unethical, biased or inaccurate outputs and corporate data leakage into the vendor’s AI model;
• Preparing for AI is the leading data storage priority in 2023, followed by cloud cost optimization;
• The majority (40%) will pursue a multi-pronged approach to manage AI risk, encompassing storage, data management and security tools;
• Organizations managing more than 10PB of data grew from 27% to 32% this year, a 19% increase;
• Half of organizations are managing 5PB or more of data, similar to 2022;
• Nearly three-quarters (73%) are spending 30% or more of IT budget on data storage and protection, measurably higher than 67% in 2022;
• The top unstructured data management challenge is moving data without disrupting users and applications (47%), followed closely by preparing for AI and cloud services (46%);
• Most (85%) say that non-IT users should have a role in managing their own data and 62% already have attained some level of user self-service for unstructured data management;
• Monitoring and alerting for capacity issues and anomalies led the pack for important future unstructured data management capabilities (44%).

Listen to the podcast:

Also available on Apple Podcasts

Watch the video:

The post Komprise President Krishna Subramanian on Generative AI and Unstructured Data appeared first on eWEEK.

Today’s companies have difficulty acquiring the right talent, as evidenced by ManpowerGroup’s 2023 Talent Shortage Survey, where 77 percent of organizations stated that it’s challenging to find skilled workers.

As a result, 71 percent of these organizations are focusing on upskilling the current workforce to meet their digital transformation needs. I’ve been covering IT trends for a long time, and talent evolution is something businesses have always had to deal with. But innovation is happening so much faster today, creating an accelerated need for skills transitions.

Cisco is among the leading tech companies making significant strides to address the skills gap, particularly in the cybersecurity and network automation sectors. Recognizing that there’s no universal solution, Cisco has adopted a comprehensive approach that spans the entirety of the learning spectrum.

“This means that no matter where someone comes in—whether they’re tech curious, trying to enter the job market, or whether they’re in the job market already—we’re making this entire learning continuum available to keep up with new technologies,” said Par Merat, VP of Cisco Learning and Certifications, during a recent analyst update on bridging the skills gap.

Cisco’s approach is multifaceted. The company works with education partners, including its learning partner ecosystem and academies. Notably, in the U.S., Cisco’s outreach extends to an impressive 45 percent of community colleges. There’s also a direct-to-learner model, allowing individuals to engage with and access their resources.

Moreover, Cisco has partnered with learning as a service (LaaS) providers like Coursera and has forged ties with government agencies. Cisco enhances this learning approach with certifications, ensuring individuals can acquire the necessary skills.

Cisco Networking Academy: The Linchpin

Spearheading Cisco’s vision for an inclusive digital future is the Cisco Networking Academy. Since its inception in 1997, it has impacted over 20 million individuals. “NetAcad” functions as a conduit between learning and job readiness. Cisco has set an ambitious goal to provide digital and cybersecurity skills training to 25 million people over the next 10 years as part of the Networking Academy program.

“We prepare individuals for careers in areas of networking, cybersecurity, and other IT roles,” said Laura Quintana, VP and General Manager of the Cisco Networking Academy. “We support countries worldwide and help them accelerate their digital agendas by addressing the skills gap. When measuring student outcomes, 3.57 million people have attributed their participation in the academy to obtaining a new job.”

Ensuring the program remains cutting-edge, the academy integrates technologies that blend traditional learning with immersive experiences, such as gamified modules, interactive sessions, and artificial intelligence (AI). One of the academy’s standout programs is the junior cybersecurity analyst pathway, offering 120 hours of in-depth training in cybersecurity.

On a related topic: The Future of Artificial Intelligence

Certifications: The Cisco Stamp of Quality

Once an individual secures a job in the tech field, the learning process doesn’t stop there. Cisco also provides accessible and certification-based training to empower professionals in their careers. Over the past three decades, the company has issued more than 4.3 million certifications.

A distinguishing feature is Cisco’s recognition as industry, rather than product, certifications. These certifications have received accreditations from renowned bodies like the American National Standards (ANSI) and the National Institute of Standards and Technologies (NIST). Cisco continually revises and adapts its certification content, reflecting the evolving job market, a recent example being the introduction of multicloud certifications.

Cisco U: The Pinnacle of Digital Training

Cisco introduced a new platform called Cisco U, emphasizing a modernized approach to digital training. The inception of this platform was a collaborative effort with the Cisco community, aiming to strike a balance between proficiency and efficiency in learning. The platform is anchored around three foundational tenets: guidance, community, and depth.

Cisco U stands out for its use of AI and machine learning (ML) to recommend specific learning paths by offering assessments that guide individuals. This ensures people access only the most relevant content for the skills they’re trying to acquire. The platform delves deep into Cisco-specific topics while also branching out to cover newer, interdisciplinary areas.

Community engagement is an essential aspect of Cisco U. Previously, the community discussions were separate from learning resources, causing a disconnect. Now, Cisco U brings these discussions directly into the learning experience. It’s a new integrated approach that fosters “vibrant conversations, tips, and informal mentoring,” Merat said.

Also see: The Pros and Cons of Deep Learning

Case Study: How North Dakota Is Bridging the Skills Gap

North Dakota, with its predominantly rural setting, faces significant challenges in technological advancement, workforce transitions, and shifts in agricultural practices. To address these, the state has taken several key steps to promote digital equity among its citizens with the help of “Skills for All,” a Cisco program that offers free online tech courses.

Recognizing the importance of comprehensive technological education, North Dakota has integrated Cisco’s educational offerings across all its institutions, ranging from pre-kindergarten all the way to Ph.D. programs. This ensures that students receive a thorough grounding in technology regardless of their academic level.

Additionally, there’s a strong push towards promoting remote work and developing skills to stay relevant in the tech landscape. Part of the vision is to expand broadband access throughout the state, ensuring that every citizen can take advantage of modern technologies like cloud connectivity and AI.

On the educational front, North Dakota launched an initiative called PK-20W, which stands for pre-kindergarten through Ph.D. and the workforce. The initiative emphasizes the importance of continuous upskilling and reskilling, offering avenues for career progression without the constant requirement of higher formal degrees.

This vision has been further cemented by recent legislation that mandates cybersecurity education as a requirement for high school graduation. Cisco plays a crucial role in PK-20W, providing resources and support, which many of the state’s educational institutions have adopted.

Upskilling the Current Workforce

Beyond just education, there’s a pressing need to adapt the current workforce’s skills, especially in the IT sector. As technologies like AI and cloud computing become more central to operations, IT professionals must be adequately trained to handle the challenges they bring. Cisco’s programs have been instrumental in offering structured training to IT staff and ensuring they remain updated in their roles.

“The digital landscape is going to continue to move forward, and we’ve got to prepare ourselves to be at the cusp of that technological change,” said Kuldip Mohanty, North Dakota Office of the Governor’s chief information officer. “That’s where the Cisco Networking Academy continues to be our strongest foundation. This collaboration extends beyond technology. It’s about fostering a culture of innovation, learning, and adaptability.”

North Dakota’s vision is technological inclusivity, where every citizen is equipped and empowered to navigate the digital age. The partnership with Cisco stands at the heart of this vision, fostering a culture of continuous learning and ensuring that the state has a skilled workforce for the future.

The topic of reskilling is something I talk to almost all IT professionals about. There are many ways to keep one’s skills current, but leveraging vendor-driven programs like Cisco offers is an excellent way to ensure the training received is current and in line with present and future technology trends.

Technology is evolving faster than ever, and those willing to continually learn new skills will have a leg up on those that don’t invest in themselves.

On a related topic: The AI Market: An Overview 

The post How Cisco is Addressing the Widening Skills Gap appeared first on eWEEK.

Privacy issues created by widespread use of artificial intelligence are a major concern, despite the fact that businesses and consumers alike are excited by the potential for AI to transform daily life. Clearly, as more and more personal data is fed into AI models, many consumers are rightfully concerned about their privacy and how their data is being used.

This guide is for these consumers to build a deeper knowledge base about AI’s privacy features. Additionally, it is a guide for business owners and leaders to better understand their customers’ concerns and how to use AI in a way that protects privacy without sacrificing functionality.

Table of Contents: AI and Privacy

• Issues with AI and Privacy
• How Is AI Data Collected?
• Solutions for AI and Privacy Concerns
• Bottom Line

Issues with AI and Privacy

Little regard for copyright and IP laws

AI models pull training data from all corners of the web. Unfortunately, many AI vendors either don’t realize or don’t care when they use someone else’s copyrighted artwork, content, or other intellectual property without their consent.

This problem grows much worse as models are trained, retrained, and fine-tuned with this data; many of today’s AI models are so complex that even their builders can’t confidently say what data is being used and who has access to it.

Also see: AI Detector Tools

Unauthorized incorporation of user data

When AI model users input their own data in the form of queries, there’s the possibility that this data will become part of the model’s future training dataset. When this happens, this data can show up as outputs to other users’ queries, which is a particularly big issue if users have input sensitive data into the system.

In a now-famous example, three different Samsung employees leaked sensitive company information to ChatGPT that could now possibly be part of ChatGPT’s training data. Many vendors, including OpenAI, are cracking down on how user inputs are incorporated into future training, but still, there’s no guarantee that sensitive data will remain secure and outside of future training sets.

Limited regulatory bodies and safeguards

Some countries and regulatory bodies are working on AI regulations and safe use policies, but no overarching standards are currently in place to hold AI vendors accountable for how they build and use artificial intelligence tools.

A number of AI vendors have already come under fire for IP violations and opaque training and data collection processes. But in most cases right now, AI vendors get to decide their own data storage, cybersecurity, and user rules without interference.

Also see: Top Generative AI Apps and Tools

Unauthorized usage of biometric data

A growing number of personal devices use facial recognition, fingerprints, voice recognition, and other biometric data security in place of more traditional forms of identity verification. Public surveillance devices also frequently use AI to scan for biometric data so individuals can be identified more quickly.

While these new biometric security tools are incredibly convenient, there’s limited regulation regarding how AI companies can use this data once it’s collected. In many cases, individuals don’t even know that their biometric data has been collected, even less that it is being stored and used for other purposes.

Covert metadata collection practices

When a user interacts with an ad, a TikTok or other social media video, or pretty much any web property, metadata from that interaction and the person’s search history and interests can be stored up for more precise content targeting in the future.

This method of metadata collection has been going on for years, but with the help of AI, more of that data can be collected and interpreted at scale, making it possible for tech companies to further target their messages at users without their knowledge of how it works. While most user sites have policies that mention these data collection practices, it’s mentioned only briefly and in the midst of other policy text, so most users don’t realize what they’ve agreed to and subject themselves and everything on their mobile devices to scrutiny.

Limited built-in security features for AI models

While some AI vendors may choose to build in baseline cybersecurity features and protections, many AI models do not have native cybersecurity safeguards in place. This makes it incredibly easy for unauthorized users and bad-faith actors to access and use other users’ data, including personal identifiable information (PII).

Extended data storage periods

Few AI vendors are transparent about how long, where, and why they store user data, and the vendors who are transparent often store data for lengthy periods of time.

For example, OpenAI’s policy says it can store user input and output data for up to 30 days “to identify abuse.” However, it’s not clear when or how the company is justified to take a closer look at individual users’ data without their knowledge.

Privacy and the Collection of AI Data

Web scraping and web crawling

Because it requires no special permissions and enables vendors to collect massive amounts of varied data, AI tools often rely on web scraping and web crawling to build training datasets.

Content is scraped from publicly available sources on the internet, including third-party websites, wikis, digital libraries, and more. In recent years, user metadata has also become a large portion of what’s collected through web scraping and crawling. This metadata is usually pulled from marketing and advertising datasets and websites with data regarding targeted audiences and what they engage with most.

User queries in AI models

When a user inputs their question or other data into an AI model, most AI models store that data for at least a few days. While that data may never be used for anything else, it has been proven that many artificial intelligence tools not only collect that data but hold onto it for future training purposes.

Biometric technology

Surveillance equipment, including security cameras, facial and fingerprint scanners, and microphones that detect human voices can all be used to collect biometric data and identify humans without their knowledge or consent.

State by state, rules are getting stricter regarding how transparent companies need to be when using this kind of technology. Yet for the most part, they can collect this data, store it, and use it without asking customers for permission.

IoT sensors and devices

Internet of Things (IoT) sensors and edge computing systems collect massive amounts of moment-by-moment data and process that data nearby in order to complete larger and quicker computational tasks. AI software often takes advantage of an IoT system’s database and collect relevant data through methods like data learning, data ingestion, secure IoT protocols and gateways, and APIs.

APIs

APIs give users an interface with different kinds of business software so they can easily collect and then integrate different kinds of data for AI analysis and training. With the right API and setup, users can collect data from CRMs, databases and data warehouses, and both cloud-based and on-premises systems.

Public records

Whether records are digitized or not, public records are often collected and incorporated into AI training sets. Information about public companies, current and historical events, criminal and immigration records, and other public information can be collected with no prior authorization required.

User surveys and questionnaires

Though this data collection method is more old-fashioned, using surveys and questionnaires is still a tried-and-true way that AI vendors collect data from their users.

Users may answer questions about what content they’re most interested in, what they need help with, how their most recent experience with a product or service was, or any other question that gives the AI a better idea about how to personalize interactions with that person in the future.

Also see: 100+ Top AI Companies

Solutions for AI and Privacy Concerns

With a handful of best practices, tools, and additional resources, your business can effectively use artificial intelligence solutions without sacrificing user privacy. To protect your most sensitive data at all stages of AI usage, follow these tips:

• Establish an appropriate use policy for AI: Internal users should know what data they can use and how and when they should use it when engaging with AI tools. This is particularly important for organizations that work with sensitive customer data, like protection health information (PHI) and payment information.
• Invest in data governance and security tools: Some of the best solutions for protecting AI tools and the rest of your attack surface include extended detection and response (XDR), data loss prevention, and threat intelligence and monitoring software. A number of data-governance-specific tools also exist to help you protect data and ensure all data use remains in compliance with relevant regulations.
• Read the fine print: AI vendors typically offer some kind of documentation that covers how their products work and the basics of how they were trained. Read this documentation carefully to identify any red flags, and if there’s something you’re not sure about or that’s unclear in their policy docs, reach out to a representative for clarification.
• Use only non-sensitive data: As a general rule, do not input your business’s or customers’ most sensitive data in any AI tool, even if it’s a custom-built or fine-tuned solution that feels private. If there’s a particular use case you want to pursue that involves sensitive data, research if there’s a way to safely complete the operation with digital twins, data anonymization, or synthetic data.

For additional tips related to cybersecurity, risk management, and ethical AI use when it comes to generative AI in particular, check out these previous best practice guides:

• Generative AI Ethics: Concerns and Solutions
• Generative AI and Cybersecurity
• Risks of Generative AI: 6 Risk Management Tips

Bottom Line: AI and Privacy Issues

AI tools present businesses and the everyday consumer with all kinds of new conveniences, ranging from task automation to guided Q&A to product design and programming. But as much as these tools can simplify our lives, they also run the risk of violating individual privacy in ways that can damage vendor reputation and consumer trust, cybersecurity, and regulatory compliance.

It takes extra effort to use AI in a responsible way that protects user privacy, but it’s well worth it when you consider how privacy violations can impact a company’s public image. Especially as this technology matures and becomes more pervasive in our daily lives, it will become crucial to follow AI laws as they’re passed and develop more specific AI use best practices that align with your organization’s culture and customers’ privacy expectations.

Read next: Best Artificial Intelligence Software

The post AI and Privacy Issues: What You Need to Know appeared first on eWEEK.

Cloud computing is having a moment. No longer an emerging trend, Gartner predicts public cloud service and user spending to grow nearly 22%, totaling more than $597 billion in 2023, up from $491 billion in 2022. Gartner also reports that 75% of organizations will utilize a cloud-driven digital transformation model by 2026.

If the massive $2 billion commitment by KPMG for investment in Microsoft Cloud and AI services is any indication, the AI arms race will fuel further investment in the cloud platforms required to store and process the large data sets required for AI applications.

As the demand for cloud services continues to rise, there is still growing concern around “cloudflation” driven by the rise in storage costs fueled by higher energy prices – pain that is being felt around the world. And there are also valid concerns about how cloud data centers contribute to higher carbon emissions. This is especially relevant given how climate change is wreaking havoc across the U.S, and the globe.

Why Cloud Will Remain Dominant

There are four key reasons why the cloud will be a dominant force, and continue to be the centerpiece of enterprise and business computing for years to come:

The generative AI-fueled arms race

The economic uncertainty may factor into IT budgets and buying decisions, but this doesn’t seem to be slowing down investment in the cloud. In fact, going forward, generative AI will be a major factor that drives many organization’s tech stack purchase decisions. Whether in the public or private cloud, generative AI is supported by large language models (LLMs) which process data in real time, requiring powerful – and highly scalable – computing power.  When you combine this point with the fact that the best AI solutions are now offered primarily as cloud services, organizations are more motivated than ever to accelerate their transition to the cloud.

Also see: Top Generative AI Apps and Tools

Affordability

Even with rising energy costs, cloud computing is the far more affordable option: And this is true for the largest Fortune 500 enterprises down to SMBs.

Committing to the on-prem model means spending more on IT systems, applications, and hardware infrastructure because it has to keep up with performance objectives, and those can change over time. This can be especially challenging for retail brands and their supply chain partners during holiday sales peaks, for example.

IT must support the increase in performance needs even if only a temporary or seasonally related. A usage-based model is always going to be less expensive than buying an on-prem system or software, which leads to contract lock-ins. As mentioned earlier, the cloud data centers do require energy and water to operate, yet companies are in a better position to meet their ESG and sustainability goals because they’re not using on-site servers to power IT infrastructure.

Application updates, security, and economies of scale

The cloud is better at regular application updates, security, and economies of scale: Despite a recent high profile cloud data breach, overall, cloud platforms offer better security because their reputations as trusted partners depend on it.

For example, the big cloud platforms have hundreds of top-notch engineers, cybersecurity professionals, and IT staff that united in the goal to keep their customer’s data safe. Microsoft, AWS, and Google also have what most small to mid-sized companies do not – economies of scale. They provide seamless application updates, which includes patching applications, and detecting and quashing vulnerabilities before breaches occur.

Companies with strapped or overworked IT staff benefit from the 24/7/365 support that global cloud platform providers offer. From a security, scale, and even a sustainability/ESG standpoint, the cloud is the clear winner.

Also see: Top Cloud Service Providers and Companies

Issues with On-Prem

On-prem isn’t a good fit for the way we work today: While some CEOs seem to be souring on the work from home trend, some form of hybrid work ecosystem will likely continue well into the foreseeable future. The changed 9-5 landscape demands high levels of accessibility for teams that may be in multiple regions, time zones or countries that on-prem software doesn’t offer.

The growth of software-as-a-service (SaaS) over the past five years, boosted heavily by the pandemic, provides the widespread application and network accessibility that companies require to be efficient, productive and remain competitive in today’s business landscape.

Bottom Line: The Cloud and Sustainability

A growing number of companies are making commitments to being better “corporate citizens” and achieving carbon neutrality, which is good news. And even better news is that carbon footprint reduction is top of mind for the Big Three: Microsoft, AWS and Google. All three have confirmed their commitment to sustainability, energy efficiency and the reduction of their carbon footprint. Both Microsoft and AWS plan to power all of their respective data centers with 100% renewable energy by 2025.

Companies don’t have to rely entirely on their providers for a “greener” cloud, they can also make a conscious effort by being more selective about how they consume cloud services, how much data they’re storing unnecessarily, and where their cloud data providers are located.

Despite some downsides to the cloud, it’s still the best strategies for the SMBs to the global Fortune 500 from a productivity, green, and affordability standpoint.

For more information, also see: Digital Transformation Guid

About the Author:

Scott Francis, Technology Evangelist at PFU America, brings more than 30 years of document imaging expertise to his position where he’s responsible for evangelizing Ricoh’s industry leading scanner technology.

The post Why Cloud Will Remain Dominant: 4 Reasons appeared first on eWEEK.

AI in the retail industry is having enormous impact, chiefly by improving workflows for workers and streamlining product selection for customers.

Some version of AI has been part of the e-tail and e-commerce experiences for several years, but particularly with the advent and growth of generative AI, digital retail experiences have become much smoother and less reliant on human intervention.

In this guide to AI’s role in retail, learn how AI is currently being used and how it impacts everyone from business leaders to hourly workers to customers.

Table of Contents: AI in Retail

• How Does AI in Retail Work?
• Benefits of Using AI in Retail
• Examples of AI in Retail
• Uses of AI in Retail
• Bottom Line: AI in Retail

How Does AI in Retail Work?

The way artificial intelligence works in the retail industry all depends on how and where business leaders choose to focus their AI efforts.

For most retail organizations, AI models are fine-tuned to act as digital retail platforms or are embedded into existing retail platforms, ERP systems, CRM software, and/or business websites. These models are trained to handle a variety of behind-the-scenes and customer-facing tasks, including helping to manage inventory, supply chain processes, customer interactions, and other features of the retail life cycle.

Retail AI deployments are trained to view every customer interaction, click, and movement of inventory as a unique data point. Often, this data is absorbed into the model’s training set and is used to further specialize and fine-tune the model’s ability to interact with humans and provide the support they need.

A variety of AI types can be used in retail, but these three broad categories cover most areas of the AI-retail process today:

• Machine learning: Machine learning algorithms are trained to recognize and act on the differences between different users and data points, including ad clicks, purchases, and customer service conversations; this type of AI is particularly useful for making predictions about overall brand sentiment, what users most want to purchase, and how purchase trends will change over time.
• Generative AI and natural language processing: Generative AI is best for building chatbots, virtual assistants, and other AIs that help customers directly by generating content in response to their queries. Generative AI models can directly answer customer questions and even make purchase suggestions based on their past purchases or preferences.
• Robotics: AI-powered robotics, which can power physical machines, may be used to scan, update, or shift inventory from warehouse to storefront; drive delivery vehicles and drones; or physically interact with customers in stores.

For more information, also see: Top Robotics Startup

Uses of AI in Retail

AI chatbots

AI chatbots with access to varying degrees of information can be embedded into customer facing e-commerce sites, social media, and other applications. Going a step beyond chatbots that rely on a few manual workflows for specific conversational topics, AI chatbots are trained on a variety of subjects — often including an organizational database or knowledge base — so they can more effectively understand and respond to diverse customer questions and requests.

Live customer service coaching

AI vendors like Nuance, Gridspace, and Zenarate don’t take over customer service interactions altogether, but instead provide live coaching and suggestions to human customer service representatives. These live suggestions are often combined with detailed customer dashboards that give reps both the tools and the language they need to have a more product-focused conversation with customers on the phone.

Customer sentiment analysis

While your organization may already be taking in and reviewing customer feedback on a regular basis, it’s often difficult to compare these sentiments across users and at scale, and even more difficult to apply that data to necessary business model improvements.

With the help of AI models, all customer feedback and interactions are automatically collected and analyzed, giving retailers a faster and better understanding of how customers feel about the brand. Particularly with large language models and generative AI models, unstructured data like text can be collected from multiple different sources, and at the same time, spammy reviews can be filtered out so they don’t muddy the waters of your analysis.

Beyond developing a better understanding of customer sentiment analysis, AI enables retailers to apply these new sentiment insights to more personalized customer experiences, including catered ads and more precise audience segmentation.

Also see: The Benefits of Generative AI

Demand forecasting and prescriptive analytics

AI-driven data analytics give users — including non-data scientists — a better understanding of all the different kinds of data at their disposal.

These models can analyze data in different formats and in large quantities, looking at historical and current purchasing metadata across different categories to more accurately forecast demand. For example, business owners may use AI analytics to learn that the sweaters they sold last year not only sold quickly but were reviewed favorably in customer reviews and conversations.

They may also learn that while the sweater sold well in the Midwest, Northeast, and most of Europe, it did not perform as well in other U.S. regions or most of Asia. Armed with this data, the retailer will know that it should bring back this product but that some markets require more stock while others require less.

For the non-data scientist, AI analytics tools are particularly effective at offering prescriptive analytics, or analytics that make recommendations for how to adjust business tactics in the future based on current data. The natural-language approach these AI tools can take help business users across departments and areas of expertise deploy this data for better results.

More on a similar topic: Generative AI and Data Analytics: Best Practices

Recommendation AI

Based on individual users’ metadata, past purchases, ad engagement, sentiment analysis, and other data-driven inputs, AI in retail can now recommend products and services to customers that they may not have otherwise considered purchasing but are likely to want.

Indeed, recommendation AI is one of the fastest-growing AI-retail areas because of how well it connects with and monetizes customer preferences.

Automated inventory management

AI-supported demand forecasting is one of the ways retailers are now more accurately predicting how much inventory they need and where and when it should be stocked. AI-driven data analytics may also help retailers determine when prices should be changed, how seasonal purchases impact inventory storage and supply chain movement, and where customer returns require more frequent inventory shifts.

In a more tangible sense, AI-driven robotics can also be used to support automated inventory management. A growing number of retail warehouses are relying on AI assistive robots to scan inventory and monitor stock levels and then restock or remove stock as needed.

Intelligent, personalized display ads

Human marketing and ad managers work behind the scenes to analyze how ads are performing and then decide what changes should be made to get more audience engagement. AIs are taking over this task on a widespread scale and are making more accurate targeting decisions, primarily because they are able to analyze a greater quantity of engagement data points more quickly.

AI software in retail also more frequently notices data patterns that humans may overlook, and can draw decisions based on past data, whether that’s updating the ad copy based on user sentiment or changing where the ad falls on a webpage based on previous heatmap data.

AI can also target ads to individuals based on their metadata, making it so they’re more likely to be interested in and engage with advertising materials. Finally, AI learn and update ads in real time, ensuring ads are always optimized for the current audience.

Also see: Generative AI Companies: Top 12 Leaders 

Simplified checkout experiences

Through a combination of biometrics and AI recognition technology, storefronts are starting to simplify the user checkout experience, including in physical stores.

For example, some stores are now allowing repeat customers to simply grab their items and walk out the door; the store’s AI recognition and scanning technology recognizes the customer and automatically charges them without requiring them to physically check out.

For customers that want a more user-friendly virtual shopping experience, many companies have added AI assistive elements to their retail apps. These features may make purchase suggestions or integrate with virtual wallets for a smoother shopping experience.

Continue reading: Generative AI: Enterprise Use Cases

Examples of AI in Retail

Each of the following enterprise companies has incorporated a different kind of AI into its workflows to simplify retail experiences:

ai.RETAIL

Accenture, the global professional services firm, recently released ai.RETAIL, an AI and data analytics platform that helps retailers better understand their data both at an individual customer and big picture level.

Its features include customizable customer-level views that show historical buying patterns and loyalty, dynamic merchandising for different customer channels, supply network digital twin development, and customer targeting capabilities. Tim Hortons, the Canada-based restaurant chain, has used ai.RETAIL and other products and services from Accenture to set up its customer loyalty app with rewards and data analytics that help them to keep customers engaged with the brand.

IBM watsonx Assistant

IBM’s watsonx Assistant is an AI virtual assistant that can be adapted to various business use cases, including customer service chatbots.

In the case of Camping World, an RV retailer, this AI solution serves as the engine for its custom-built AI chatbot, Arvee. The virtual agent has been designed to handle more straightforward customer service engagements and triage more complex conversations to human customer service representatives through dynamic routing.

It is also able to save detailed information about customer inquiries that occur after-hours, allowing employees to step in and re-engage those customers as necessary once they’re back on the clock. Using IBM’s AI virtual assistant has freed up more time for the company’s employees and also helps them to focus on the most pressing and challenging customer inquiries.

Amazon One and Amazon Go stores

Amazon One and Amazon Go are two AI-driven solutions Amazon has built to simplify the customer-side of each retail interaction with a shopping experience it calls “Just Walk Out” shopping.

With Amazon One, a customer’s payment information is linked to their palm, and AI and machine learning are used to identify that palm and charge the appropriate person’s account when they leave an Amazon Go or other participating store with a purchase. This means that users can quickly pick out their items and leave without going through a traditional checkout process, and Amazon’s AI sensors quickly detect when an item is picked up or put back on the shelf.

Benefits of Using AI in Retail

Many retailers fear that using AI in their business model will cause them to lose their personal touch, but so far, it has offered early adopters a range of benefits — including helping them to create a more customer-friendly experience:

• Emphasis on customer experience: Ads are targeted to what users actually want, chatbots can more clearly answer user questions on a customer’s schedule, and AI-driven apps give users access to new types of shopping experiences that fit their preferences. Although there may be less human-to-human contact in retail as AI is adopted, customers are still on the receiving end of a customer-first experience.
• More automation opportunities: Artificial intelligence can automate conversational workflows, inventory and supply chain management, and other repetitive retail tasks that have traditionally required a human touch. This reduces the chance for human error and frees up employee time for more strategic tasks, both of which can lead to higher levels of organizational productivity.
• Lessened impact of employee-user error: AI often monitors the entire supply chain and inventory management lifecycle and can identify an error as soon as it occurs. This makes it easier to mitigate stocking and shipping errors before they lead to unhappy customers or inventory shortages.
• Optimized digital marketing and analytics: AI-powered analytics tools democratize the analytics process with natural language inputs, contextualized explanations, and more detailed and accurate predictive analytics that are useful to marketing and sales teams. These tools can also analyze greater quantities and different types of data than most traditional marketing analytics tools can.
• Fewer human touch points required: As AI takes over different customer service and inventory management touchpoints, retailers can reduce their staff or focus their attention on more strategic tasks. Especially as more and more of the workforce moves away from retail and service-based industries, these AIs will help to fill in a production gap with limited retraining and recruitment requirements.

Bottom Line: AI in Retail

Retailers are getting creative and finding all kinds of ways to incorporate AI into the work they do. This ranges from allowing AI to directly interact with or indirectly influence customer interactions, restock and monitor inventory across distributed sites, or give business leaders a more detailed glimpse into current performance data and future projections.

If AI is implemented thoughtfully and with appropriate usage policies and safeguards in place, it can benefit both your employees and customers with automation, personalization, and hands-off features that improve the overall retail experience.

Read next: Top 9 Generative AI Applications and Tools

The post AI in Retail: What You Need to Know appeared first on eWEEK.

I spoke with Chat Kapoor, CEO of Datastax, about what he refers to as the relentless pace of generative AI’s growth; we also discussed data market trends that influence large language models.

Among the topics we discussed: 

• How do you see the rise of generative AI affecting enterprise workflows?
• What’s your take on the current data market? Will companies need their own LLM? That seems cost prohibitive for most companies.
• You’ve mentioned the role of vector databases in making companies more competitive with AI. How so?
• The future of generative AI in the enterprise? What major milestones do you see in the next 12-24 months?

Listen to the podcast:

Also available on Apple Podcasts

Watch the video:

The post Datastax CEO Chet Kapoor on the Data Market and AI appeared first on eWEEK.

I spoke with Peter Pezaris, Chief Strategy and Design Officer, New Relic, about a new report about Observability in the enterprise. A huge new rising trend: generative AI.

Among the topics we discussed: 

• Before we get to the report, what trends do you see in the deployment of observability in the enterprise? Are companies struggling with observability in any way?
• Let’s look at a few key data points from the 2023 Observability report.
• How is New Relic addressing the observability needs of its clients? What’s the New Relic take on the market?
• The future of observability in the enterprise? What are some key milestones we can expect in the years ahead?

Listen to the podcast:

Also available on Apple Podcasts

Watch the video:

The post New Relic’s Peter Pezaris on Observability Trends appeared first on eWEEK.

Generative AI chatbots are a major step forward in conversational AI. These chatbots are powered by large language models (LLMs) that can generate human-quality text, translate languages, write creative content, and provide informative answers to your questions.

An ever-growing list of generative AI chatbots are now entering the market, but not all chatbots are created equal. We analyzed the best generative AI chatbots to help you determine the best conversational AI app for your business.

Here are our picks for the top generative AI chatbot software. Jump to:

• Top generative AI chatbot software comparison chart
• Bing Chat Enterprise: Best for organizations in the Microsoft ecosystem
• ChatGPT: Best for versatility and advanced generative AI features
• OpenAI Playground: Best for customizability
• Perplexity AI: Best for finding information on the Internet
• YouChat: Best for students and researchers
• ​​Chatsonic: Best for individuals in the creative industries
• Google Bard: Best for brainstorming ideas
• Socrates.ai: Best for internal knowledge-base management
• HuggingChat: Best for developers
• Jasper: Best for marketing and sales team
• Key features of generative AI chatbot software
• How do I choose the best generative AI chatbot software for my business?
• How we evaluated the best generative AI software
• Frequently Asked Questions (FAQ)
• Bottom Line: Best AI Chatbots

Top Generative AI Chatbot Software: Comparison Chart

We compared the key features of the top generative AI chatbot software to help you determine the best option for your company:

Please note: All ratings are based on a range of 1 to 5, with 5 being the best option.

Bing Chat Enterprise: Best chatbot for organizations in the Microsoft ecosystem

Overall rating: 3.13

• Cost: 3.95
• Feature Set: 3.5
• Ease of Use: 1.75
• Quality of Output: 2.5
• Support: 4

Microsoft generative AI tool Bing Chat Enterprise uses GPT- 4 – a large language model – to generate natural language responses to users’ queries. Bing Chat Enterprise has three conversation styles: Creative, balanced and precise. These styles help users set the tone for the expected response to their query. Bing Chat Enterprise is available in 160 regions, and you can access it via Bing.com/Chat, Microsoft Edge Sidebar or Windows Copilot Sidebar.

Pricing

Bing AI chat is available at no additional cost for customers who are licensed for Microsoft 365 E3, E5, Business Standard, Business Premium, or A3 or A5 for faculty. If you don’t have those licenses, you can purchase Bing AI as a standalone tool for $5 monthly.

Features

• A versatile mix of conversation styles for different enterprise AI use cases.
• Bing’s citation capability allows you to fact-check its output.
• It allows users to upload an image and it can also generate an image for you.

Pros

• Bing’s citation can enable you to continue your research on your own.
• The user interface is visually appealing and easy to navigate.
• It has an AI built-in image generator.

Cons

• While trying the tool on non-edge browser, I got an error message “You’ve reached the conversation limit for today. Sign in to continue your chat.” I found that other users experience this too, even though they are signed in or haven’t used the tool before. To resolve this issue, I downloaded Microsoft Edge.
• Limited to 30 responses per conversation.

Also see: Top Generative AI Apps and Tools

ChatGPT: Best chatbot for versatility and advanced generative AI features

Overall rating: 4.51

• Cost: 4.15
• Feature Set: 4.4
• Ease of Use: 5
• Quality of Output: 3.75
• Support: 5

Developed by OpenAI as part of the GPT (Generative Pre-trained Transformer) series of models, ChatGPT is a natural language processing tool designed to engage in human-like conversations with users. The platform can perform NLP tasks, such as answering questions, providing recommendations, summarizing text, and translating languages. Aside from content generation, developers can also use ChatGPT to assist with coding tasks, including code generation, debugging help, and answering programming-related questions.

Pricing

• Free: You can use the free plan at no cost.
• ChatGPT Plus: $20 per month.
• Enterprise plan: Custom quotes.

Features

• Analytics dashboard to understand usage for enterprise users.
• Multilingual features: It can understand and generate text in multiple languages.
• Contextual understanding of extended conversation.

Pros

• Data encryption at rest (AES-256) and in transit (TLS 1.2+).
• ChatGPT can assist in completing sentences or paragraphs for users.
• Explain complex concepts.

Cons

• The free plan is limited to GPT-3.5.
• Platform knowledge is limited to 2021.
• Biases and inaccurate information.

For more information: Full ChatGPT review.

OpenAI Playground: Best chatbot for customizability

Overall rating: 3.0

• Cost: 4.15
• Feature Set: 5
• Ease of Use: 0.55
• Quality of Output: 2.5
• Support: 2

OpenAI Playground was designed by the same AI research company that created ChatGPT. It is a web-based environment allowing users to experiment with different OpenAI models, including GPT-4, GPT-3.5 Turbo, and others. OpenAI Playground is suitable for advanced users looking for a customizable generative AI chatbot model that they can easily fine-tune to suit their needs.

Pricing

OpenAI Playground is priced per 1,000 tokens (1,000 tokens equals to 750 words). Pricing for this tool varies based on your chosen model. New users can get $5 in free credit to use for their first three months.

GPT-4 models

• 8K context: Input costs $0.03 per 1K tokens while output costs $0.06 per 1K tokens.
• 32K context: Input costs $0.06 per 1K tokens while output costs $0.12 per 1K tokens.

GPT-3.5 Turbo

• 4K context: Input costs $0.0015 per 1K tokens, while output costs $0.002 per 1K tokens.
• 16K context: Input costs $0.003  per 1K tokens, while output costs $0.004 per 1K tokens.

Features

• Text and completion summarization capability.
• Multiple options for model selection.
• Language translation capability.
• Sentiment analysis.

Pros

• Available on the web and app for Android and iOS devices.
• Highly customizable.

Cons

• Limited creativity.
• Privacy concerns.

Also see: The Benefits of Generative AI 

Perplexity AI: Best chatbot for finding information on the Internet

Overall rating: 4.19

• Cost: 4.15
• Feature Set: 4.4
• Ease of Use: 5
• Quality of Output: 5
• Support: 2

Perplexity AI is a generative AI chatbot/search/answer engine that allows users to express queries in natural language​​ and provides answer based on information gathered from various sources on the web. When you ask a question on Perplexity AI, it provides the answer to your query and suggests related follow-up questions, you can either select from the suggested related questions or type yours in the text field.

Perplexity AI’s Copilot feature can guide users through the search process with interactive follow-up questions, multiple searches, and summarized results – this capability is helpful when researching complex topics. However, it’s limited to five searches every four hours for free plan users and up to 300 for paid users.

Pricing

• Free: Available at no cost with limited features.
• Pro: It costs $20 per month or $200 per year.

Features

• Support Claude-2 and GPT-4.
• Support 30 languages, including English, Bengali, Danish and more.
• Text and PDF file upload capability.

Pros

• Cites information sources.
• Mobile app for iOS and Android device users.

Cons

• File upload is limited to three per day for free users.
• Customer support can be improved.

YouChat: Best chatbot for students and researchers

Overall rating: 3.64/5

• Cost: 5/5
• Feature Set: 2.9/5
• Ease of Use: 5/5
• Quality of Output: 3.75/5
• Support: 1/5

Designed by You.com, YouChat is an AI-powered generative chatbot that can summarize text, write code, suggest ideas, compose emails and answer general questions based on information available on the web. It also cites its information source, making it easy to fact-check the chatbot’s answers to your queries. YouChat combines various elements in its search results, including images, videos, news, maps, social, code, and search engine results on the subject.

Pricing

• Free: Available at no cost for up to 10 AI writing generations.
• YouPro: $9.99 per month.
• YouPro for Education: $6.99 per month.

Features

• Code generator.
• Multimodal search capability.
• Supports social media profiles.

Pros

• Chrome extension available.
• It provides interactive search results with various elements.

Cons

• Contextual understanding can be improved.
• It sometimes displays outdated links or information.

Also see: Generative AI Examples

Chatsonic: Best chatbot for individuals in the creative industries

Overall rating: 4.03

• Cost: 4.55
• Feature Set: 4.4
• Ease of Use: 5
• Quality of Output: 2.5
• Support: 2

Trained and powered by Google Search to converse with users based on current events. Chatsonic advertises itself as a ChatGPT alternative. The AI chatbot is a product of Writesonic, an AI platform for content creation. When using Chatsonic, you can toggle on the “Include latest Google data” button to add real-time trending information.

Chatsonic has a Generate AI Art feature that enables it to generate digital AI artwork for users’ consumption. It also has a personality functionality that allows users to select from a list of personalities such as interviewer, math teacher, motivational coach, English translator and more – this allows the tool to provide tailored responses based on their chosen persona.

Pricing

• Free trial: Up to 10,000 words per month.
• Unlimited: $20 per month when billed monthly or $16 per month when billed yearly – limited to one user.
• Business: $19 per month when billed month or $12.67 per month when billed annually – for up to 200,000 words for GPT 3.5 or  33,3333 words for GPT 4.0 – limited to one user. With this plan, you can increase your word count and number of users at an additional cost.
• Enterprise: Custom quote.

Features

• It has a Sonic Editor (Google Docs-like Editor).
• Integration with third-party apps.
• Offers over 100 AI templates.

Pros

• Support about 25 languages.
• Landing page generator capability.
• Browser extension available.

Cons

• Priority support is limited to business and enterprise plan users.
• Free plan lacks email support.

On a related topic: What is Generative AI?

Google Bard: Best chatbot for brainstorming ideas

Overall rating: 4.28

• Cost: 5
• Feature Set: 5
• Ease of Use: 4.5
• Quality of Output: 5
• Support: 1

Google Bard is an AI-powered LLM chatbot built on the PaLM2 (Pathways Language Model, version 2) AI model. You can export your Google Bard conversation to Google Docs or Draft in Gmail – the platform also allows you to create a shareable public link you can send to a third party. Google Bard is still relatively new and is being actively developed.

Pricing

Google Bard is free to use.

Features

• Ability to export conversation.
• Its “Google it” function directs you to a search engine, providing more information on your query.
• It can generate code in at least 20 programming languages.

Pros

• It has a vast knowledge-base.
• Cites sources of information.

Cons

• Users reported that Google Bard is slow to respond, especially when users ask complex questions.
• It sometimes gives inaccurate or incomplete information.

Also see: ChatGPT vs. Google Bard: Generative AI Comparison 

Socrates.ai: Best chatbot for internal knowledge-base management

Overall rating: 1.20

• Cost: 0
• Feature Set: 3
• Ease of Use: 0
• Quality of Output: 0
• Support: 2

Socrates.ai is an artificial intelligence platform that provides businesses with conversational AI solutions. It enables companies to create and deploy conversational agents that can interact with users naturally. It can be integrated into various channels such as websites, mobile apps, and messaging platforms to enhance user experience and support automation.

Pricing

Available upon request.

Features

• Integration with third-party enterprise apps.
• It has a database of about 1000 of employee questions.
•  Supports natural language processing and understanding.
• Bot-to-bot conversations.

Pros

• Flexible deployment – SaaS or On-Premise.
• It offers a pre-built and pre-tested content that includes a glossary of terms, abbreviations, and acronyms that can be customized for your company.

Cons

• Lacks transparent pricing.
• Support can be improved.

Also see: Generative AI Companies: Top 12 Leaders

HuggingChat: Best chatbot for developers

Overall rating: 3.33

• Cost: 5
• Feature Set: 2.9
• Ease of Use: 4.75
• Quality of Output: 1.25
• Support: 1

Developed by Hugging Face, HuggingChat is a chatbot based on the Open Assistant Conversational AI Model. It uses NLP and ML algorithms to interact with users and can generate answers to questions, write essays, write code, translate text, and construct emails. The platform has been trained on a large dataset of diverse conversations and can learn from new interactions.

Pricing

HuggingChat is free to use. You can also subscribe to the pro plan at $9 per month to get early access to new features. They also offer an enterprise hub plan at $20 per user per month.

Features

• It has a web search functionality that complements its response with web content.
• Multilingual support.
• It can debug and write code and create Excel formulas.

Pros

• Easy to use.
• Highly customizable – the Hugging Face platform allows developers to create their custom intents, entities, and actions.

Cons

• Knowledge-base not up to date.
• The platform sometimes gives an incomplete answer.

Also see: Generative AI Startups

Jasper: Best chatbot for marketing and sales team

Overall rating: 4.23

• Cost: 2.65
• Feature Set: 5
• Ease of Use: 5
• Quality of Output: 5
• Support: 3

Jasper generative AI chatbot can be trained on your brand voice to interact with your customer in a personalized manner. Jasper partners with OpenAI and uses GPT 3.5 and GPT 4 language models and their proprietary AI engine. The company also sources from other models such as Neo X, T5, and Bloom. One of Jasper’s USP (unique selling points) is its brand voice functionality, which allows teams and organizations to create on-brand content.

Pricing

• Creator: It costs $49 per month when billed monthly and $39 per month when billed annually.
• Teams: It costs $125 per month when billed monthly and $99 per month when billed annually.
• Business: Custom quotes.

Features

• Browser extension available.
• Support up to 30 languages.
• Integrates with  Zapier, Webflow, Make, Google Sheets, and other tools.
• Collaboration capability.

Pros

• Provides up-to-date information and cites sources.
• Offers over 50 built-in templates.
• Its user management capability allows you to control who can access documents in your workspace with user-specific permission settings to make private, view, edit, or be available to everyone.

Cons

• Some users reported that Jasper is somewhat pricey.
• Users may experience incomplete sentences when using Focus mode with Jasper.

Key Features of Generative AI Chatbot Chatbots

Natural Language Processing

Natural language processing is an important feature of a generative AI chatbot. NLP enables the AI chatbot to understand and interpret natural language input from users, allowing them to have more human-like conversations. With NLP capabilities, generative AI chatbots can recognize context, intent, and entities within the conversation.

Context Understanding

Chatbot’s ability to comprehend and retain context during conversations enables a more seamless and human-like conversation flow. A quality artificial intelligence chatbot can maintain context and remember previous interactions, providing more personalized and relevant responses based on the conversation history. This enables chatbots to provide more coherent and relevant replies.

Personalization

When shopping for a generative AI chatbot software, customization and personalization capabilities are important factors to consider, as it enables the tool to tailor responses based on user preferences and history. ChatGPT, for instance, allows businesses to train and fine-tune chatbots to align with their brand, industry-specific terminology, and user preferences.

Multilingual support

AI chatbot’s ability to communicate in multiple languages makes it appealing to global audiences. This functionality also allows the chatbot to translate text from one language to another.

Also see: Best Artificial Intelligence Software 2023

How to Choose the Best Generative AI Chatbot For Your Business

The best generative AI chatbot for your company serves your business’s needs and balances quality service with moderately expensive – or affordable pricing – based on what  works with your budget, plus the necessary AI features you need for your operations.

Organizations in the Microsoft ecosystem may find Bing Chat Enterprise beneficial, as it works better on Edge browser. ChatGPT does not cite sources, but it is one of the most versatile and creative AI chatbots. Google Bard cites sources and provides up-to-date information, but its response time is sometimes slow. Bing Chat and Chatsonic can generate AI images as part of the answer to your query.

What seems like positives to you may be negatives to another user and vice versa. The best tool for your business is unique to you – conduct your own research, identify your goals, and shop for a tool that offers features and capabilities that meet your requirements.

How We Evaluated the Best Generative AI Chatbots

Cost – 20%

We reviewed each AI chatbot pricing model and available plans, plus the availability of a free trial to test out the platform. Our research found that some platforms are completely free while some offer both free and paid plans – a tool like Google Bard gives you access to all its features for free, ChatGPT has a free plan with access to GPT 3.5 capabilities, while GPT 4 requires a monthly subscription. On the other hand, Jasper is a completely paid chatbot offering a seven-day free trial.

Features set – 30%

Features carry the most weight (30%) in our evaluation process. We evaluated various capabilities offered by each generative AI software, including multi-language support, the ability to accept spoken word input, the programmability of the solution, the kind of users it is built for and customization options.

Ease of use – 25%

We assessed each generative AI software’s user interface and overall user experience. This included evaluating the ease of installation, setup process, and navigation within the platform. A well-designed and intuitive interface with clear documentation, support materials and the AI chatbot response time contributed to a higher score in this category.

Quality of output – 10%

To determine the output quality generated by the AI chatbot software, we analyzed the accuracy of responses, coherence in conversation flow, and ability to understand and respond appropriately to user inputs. We selected our top solutions based on their ability to produce high-quality and contextually relevant responses consistently.

Support – 15%

Our analysis also considered the level of support provided by the AI software provider. We assessed the availability and responsiveness of customer support, including customer service hours, email support, live chat support and knowledge base.

AI Chatbots: Frequently Asked Questions (FAQs)

We answered the most commonly asked questions about AI chatbots below.

What features should businesses look for in AI chatbots?

Key features to look for in AI chatbots include NLP capabilities, contextual understanding, multi-language support, pre-trained knowledge and conversation flow management. It is also important to look for a tool with a high accuracy rating, even if the questions asked are complex or open-ended.

Why are AI chatbots important for businesses?

AI chatbots are important to businesses because they enhance customer experience and provide various operational benefits such as improved customer experience, personalized experiences, cost reduction, and increased productivity.

How can AI chatbots enhance customer support and engagement?

AI chatbots can boost customer support by providing 24/7 support, answering common questions, and personalizing interaction based on customer preferences (for instance, multilingual AI chatbots can communicate in multiple languages, enabling businesses to assist customers from different regions.

Bottom Line: Top AI Chatbots

Determining the “best” generative AI chatbot software of 2023 can be subjective, as it largely depends on the business’s specific needs and objectives. The landscape of AI chatbot software is continuously evolving, and new chatbot entrants may offer innovative features and improvements over existing solutions. Therefore, the best chatbot for your business will vary based on factors such as industry, use case, budget, and desired features – there is no “one size fits all” solution.

Read next: 100+ Top AI Companies 2023

The post 10 Best AI Chatbots 2023 appeared first on eWEEK.

I spoke with Lori Rosano, Head of Public Cloud, SAP North America, about what’s driving public cloud growth in 2023, and her forecasts for key cloud developments.

Among the topics we discussed:

• What major trends do you see driving public cloud usage this year? If a company has not fully embraced the cloud by 2023, what’s holding them back?
• What’s a common challenge you see with public cloud clients? Some thoughts on addressing this challenge?
• How is SAP addressing the public cloud needs of its clients? Is SAP best suited to large enterprise clients?
• The future of public cloud? It seems like one of its new – and most powerful – drivers is its ability to facilitate/enable artificial intelligence. Thoughts?

Listen to the podcast:

Also available on Apple Podcasts

Watch the video:

The post SAP’s Lori Rosano on the Future of Public Cloud appeared first on eWEEK.

Over the past decade, the role of the network has changed: What was once viewed as a non-differentiated tactical resource that got very little attention from business leaders is now considered a strategic asset.

An interesting data point from ZK Research is that 68% of business leaders consider the network important to business transformation. This is because organizations have become more distributed and dynamic, putting the network at the center of digital transformation strategies.

Also see: Top Digital Transformation Companies

WAN Evolution is Necessary for Digital Transformation

The network, particularly the WAN, has not evolved in the better part of thirty years. In fact, the architecture most WANs are designed with is essentially the same as when I was an IT pro over two decades ago. This is what’s driving the high amounts of interest in software-defined WANs (SD-WANs). Despite the momentum, finding good case studies of real-world deployments has been challenging.

At the VMware Explore 2023, VMware hosted a breakout session with two technical implementation team members from Providence, a healthcare organization with 51 hospitals and 829 clinics in a wide variety of states around the country. 

Providence Used SD-WAN to Redesign its Network

The case study senior principal network engineer Charlie Hagen and principal network engineer Conor McCutcheon were at the VMware event to discuss “Redesigning a Mission-Critical WAN to Drive Business.”

McCutcheon described his organization’s journey. “We provide life-saving care,” he said. “Underpinning all of that healthcare delivery is our network. A patient having a stroke or heart attack does not want to wait for the network to work for them to receive that care.”

McCutcheon said that the state of the Providence network was driven by organic growth and expansion over many years. This resulted in a fractured WAN topology with different regional networks that didn’t use a common standard.

In my experience, this is, by far, the norm versus the exception, leaving most companies with a messy, unmanageable WAN environment.

WAN Evolution Required a Rethink of the Network

McCutcheon said that Providence took these actions:

• Consolidated to a single routing design enterprise-wide
• Standardized on a single routing platform
• Mediated overlapping IP space
• Unified disparate routing domains
• Set enterprise-wide standards

The consolidation fixed many issues, but selecting a single national carrier was challenging. With a national carrier, you have to take what they offer. That carrier might subcontract services to the local exchange, leading to finger-pointing when something goes wrong, which increases mean-time-to-recovery and can slow down new services or upgrades.

“Finally, we didn’t always see eye-to-eye with our national carriers on [technological] diversity,” McCutcheon said. “There were numerous occasions where incidents like fiber-seeking backhoes or carrier maintenance took both circuits down at a site even though we ordered diversity for that site.”

Charlie Hagan joined the discussion and said Providence decided to return to the table and reexamine the requirements. They wanted to develop the best WAN in healthcare and embrace change rather than depending on yesterday’s architecture and approaches.

SD-WAN Provides Superior Application Experience 

“We wanted to deliver the best application experience because our caregivers experience our network through the applications that consume it,” Hagan said. “We wanted the WAN to enable the business to be agile in response to new opportunities and demands, including absorbing bandwidth-intensive needs of new medical equipment rather than being the bottleneck.”

Being able to shift the network because of unforeseen events (maybe a global pandemic?) was a critical requirement.

Hagan and McCutcheon decided to select an SD-WAN to enable the following advantages:

• Dynamic remediation of impaired circuits through error correction and real-time detection and failover of impaired circuits.
• Abstraction of the transport from the routing to enable integration of additional circuit types like internet connectivity into an enterprise connectivity model without having a separate solution.
• Simplified management of the entire platform from a single pane of glass.
• Standardization of the configuration and design.

“As Yogi Berra said, ‘When you come to a fork on the road, take it,'” McCutcheon said. “Our proposal was nothing less than a reinvention of our WAN.” So, with stakes so high, McCutcheon said the selection had to be correct. Providence asked a third party to build a test environment.

For more information, also see: Digital Transformation Guid

Selecting the SD-WAN Solution

“Several platform vendors participated and underwent an extensive test plan to ensure that they met our requirements and that we understood their behavior under load and failure conditions,” he said.

McCutcheon then shared some criteria they used to select their next SD-WAN solution, including dynamic path selection, zero-touch provisioning, orchestration and automation, link impairment detection and remediation, and interoperability with the legacy network.

“So as a result of this extensive testing, and because we’re here at VMware Explore, you shouldn’t be surprised to learn that VMware was our selected SD-WAN vendor,” McCutcheon said. “Since we’ve made that selection, we’ve finished the final design of our network integrating SD-WAN and begun to implement.”

As a result, McCutcheon said that the Providence WAN is rapidly becoming a strategic asset that enables the organization to be agile when new opportunities and demands arise. Now, Providence can move healthcare out of hospitals and into its network of ambulatory clinics.

No transition like this is flawless. In fact, Providence experienced an issue right off the bat. “During our very first cutover, we experienced an unplanned failover event during a morning executive briefing,” Hagan said. “This was a user, not a platform error. But after the event, we were pleased to be able to surprise the executives with the news that we’d had a failover. There was no perceived impact during their call. Since that time, we continue to see the same excellent responsiveness to circuit issues. The VMware SD-WAN platform provides detailed reportable telemetry on the performance of our network.”

Providence recently had another incident where they used telemetry within the SD-WAN to isolate an issue. “Because our internet upstream at our data center was the endpoint for multiple tunnels for multiple locations, we were able to identify that only some of those tunnels were impacted by a carrier issue unidirectionally,” McCutcheon said. “In short, we were able to feed that to our upstream provider to identify that there was a peering issue upstream of them. It was all available to us within the orchestrator, and it was enlightening.”

Maybe more important, Providence can now use telemetry to hold their carriers accountable and to identify the graded circuit performance despite the lack of impact on caregivers. We throw around terms like mission-critical all the time, which, in the grand scheme of things, isn’t really accurate most of the time.

But, at healthcare organizations like Providence, those words ring true. They are, quite literally, saving lives. It’s nice to see the thinking that Providence put into their move to an SD-WAN. The willingness to rethink WAN operations and break the status quo is something all companies should consider.

The post How Providence Healthcare Redesigned its WAN for Digital Transformation appeared first on eWEEK.

Every week, I have many calls and meetings—Zoom, Cisco Webex, Google Meet, Skype (believe it or not), and the dreaded Microsoft Teams. Whenever I get an invite for a Teams call, I wonder what will happen when I click the link.

Teams can be unpredictable. Say what you will about the other players, but they just work.

Zoom started as a standalone product—and if it didn’t work, the company would’ve gone under. Webex was one of the earliest players, so they know how to make meetings work. Google Meet always feels strange, but it’s generally competent, although essentially a “me too” product. Skype is, well, Skype. It’s nothing fancy—kind of AOL Messenger on steroids—but it works.

Poor User Experience is the Teams Challenge

The trouble with Teams is that, in an effort to integrate all its products, Microsoft appears to have forgotten about the experience of actually using the individual product. Most people don’t care about integration—or, more correctly, users don’t care about integration if the product itself doesn’t work.

The Teams experience, at least for me and many users I have talked to, has not been good. Most times I go to use it, I get stuck in some kind of login issue. On more than one occasion, almost as a throwback to the late 90s or early 2000s, I’ve resorted to just using the dial-in numbers.

If I can solve login, I have to change the settings for backgrounds, etc., as I’m getting set to enter a meeting. Meanwhile, the clock is ticking, and I’m late for a meeting.

One of the most significant issues with Teams is guest access. If I host a meeting or call, I’ll use Zoom, so I often join Teams as a guest.

The experience as a guest is inconsistent. Some features, like chat, work on some calls and not others. It also creates multiple user IDs, which can create confusion for end users. This is why many organizations use Teams for internal meetings but a product like Zoom for external. Guest access is better in the latest release but it’s still not on par with other vendors.

Also see: Digital Transformation Guide

The EU and Unbundling

Once inside Teams, things work all right. Yet the experience, compared to Zoom or Webex, is severely lacking. But there may be hope on the horizon—in the unlikely guise of EU regulators. You may have read that the EU has prompted Microsoft to unbundle Teams.

One news article I read about the unbundling said Teams was considered the crown jewel of the Office 365 suite. I’m not sure about that, but it’s apparent that Microsoft is unhappy about unbundling. From a business standpoint, no one would be happy about this. Government regulators interfering in the way a company operates is generally abhorrent.

Looking back, Microsoft wasn’t happy about making it easier to uninstall Explorer after wrangling with the Justice Department in the late ’90s and early 2000s. But they survived allowing non-Microsoft software into its walled garden and grew into a trillion-plus dollar company—despite years of feckless management under Steve Ballmer. 

For users of Teams in the EU, unbundling could be a good thing if Microsoft approaches it correctly. Thinking about the user experience for Teams itself rather than the overall integration should help the company address the usability issues.

The ability to link to a doc from SharePoint is probably not a high priority for people just trying to have a seamless meeting. But those integrations should be relatively easy to maintain. One big step the company could take in improving usability is to acquire a company like Box. One might think that Box and Microsoft OneDrive are overlapping products, but the reality is that Box improves the Microsoft user experience better than Microsoft’s own product.

I know many people think I’m anti-Teams but, in reality, I’m anti-bad user experience. I want to like Teams. I really do. I hope that the company will use the unbundling to free it from those shackles everywhere and focus its usability experts on creating a great product that is easy to use and just works.

Microsoft Needs to Prioritize Usability 

I might be shouting into the void, but Microsoft product engineers might want to rethink how they approach Teams. It’s a simple thought: Prioritize the core functionality that users need or want. Then, approach that development from the inside out, like Zoom did. Think about enabling great meetings and messaging. Worry about everything else later.

The challenge with Microsoft is that the company appears to prioritize locking customers into license bundles over everything else, including usability. By making the job of procuring the product so frictionless, it can use that audience to push substandard products on corporate workers. This is why so many workers will say something like, “I really don’t like using Teams, but the IT department made the decision.”

The danger that unbundling creates is that it removes that lock and creates an opening for better products. Proof of that is how fast Google managed to have Chrome usurp Internet Explorer’s once dominant position.

If you’re a customer of Teams in one of the countries where it will be unbundled, do your due diligence and look at Zoom, RingCentral, Webex, 8×8, Avaya, or any of the other many products and use Box as the solution to integrate data across them. You’ll have happier customers and likely save yourself some money.

Read next: Top Digital Transformation Companies

The post Challenges of Microsoft Teams: User Experience and Unbundling appeared first on eWEEK.

Generative AI has quickly proven itself as a valuable asset to businesses’ workflows and operations. This is true whether a business uses ChatGPT Enterprise or another of the growing list of generative AI tools and apps.

Generative AI can support staffers in managing their existing task loads and, in some cases, these models can be trained to take on entirely new tasks and types of work. In all of these cases, generative AI is helping businesses streamline and automate their processes in repeatable and scalable ways that contribute to business growth goals.

Depending on your business priorities and requirements, generative AI can support your operations in a variety of ways. Read on to learn some of the most common ways generative AI is used to improve businesses today. Also important, learn about generative AI usage best practices that will help you achieve new levels of business success.

Also see: ChatGPT Enterprise: AI for Business

Table of Contents: How Generative AI Can Support Business Operations

• Top 7 Ways to Use Generative AI in Your Company
• 5 Benefits of Leveraging Generative AI in Your Business
• 5 Challenges of Using Generative AI
• What Are Generative AI Best Practices for Business Usage?
• Bottom Line: How to Use Generative AI in Your Business

Top 7 Ways to Use Generative AI in Your Company

Generative AI models are being used for a variety of industry-specific enterprise use cases that range from simple customer service support and coaching to more complex tasks like medical imaging and drug discovery.

But generative AI can also be trained to handle more generic business tasks that are relevant to all businesses, regardless of industry or size. Below are seven of the most useful ways business leaders can incorporate generative AI into their business today:

1) Smart, Secure Data Analytics

Generative-AI-powered data analytics solutions make data analysis smarter, speedier, more scalable, and more secure. Depending on the model type and size you select, you can use generative AI to generate synthetic data for more secure and compliant data analysis and QA testing practices.

You can also use generative AI to democratize data analytics in two important ways:

• Generative AI models allow users to create narrativized explanations of their data, making it easier for non-data scientists to understand the data they have in front of them.
• Generative AI assistants are being embedded and used in common data tools like Power BI, which enables less technical users to comprehend complex datasets and create accurate data visualizations and reports.

Example solutions: SparkBeyond Discovery, Dremio, Narrative BI, Power BI Copilot

2) Customer Service and User Experience

Most businesses have a customer service component that could be improved with more consistent training and customer-first communication and designs.

A variety of generative AI tools are springing up to mentor your existing customer service agents and, in some cases, replace them with chatbot, voice generation technology, and AI-powered search engines that are designed with customer queries and natural language requirements in mind.

In more specialized industries, like insurance, generative AI can also be used to better assess risk and achieve optimal outcomes for customers’ plan rates and claims.

Example solutions: Gridspace, IBM Watson Assistant, UltimateGPT, Zendesk Advanced AI, Forethought SupportGPT

3) Assistive Coding and Product Design

Regardless of your tech teams’ levels of expertise, generative AI can be incorporated to support coding and quality assurance tests for digital product design. Generative AI models like ChatGPT can fix bugs, generate test code, and write documentation for programs.

Perhaps most significant, generative AI models can help developers of all skill levels solve problems; even non-technical team players can generate code through natural language queries.

Example solutions: Code Snippets AI, ChatGPT, Google Bard, Tabnine, MOSTLY AI, Stability AI

4) Content Creation

Generative AI models have already proven their ability to quickly generate natural language content affordably and at scale, which has made these models particularly enticing for organizations that want to outsource content writing.

Several generative AI tools are able to write blog and social media content based on simple prompts. Others are designed to create more specific content, like legal documents or marketing materials. However, the area of generative AI content generation that is growing most quickly is multimedia content, with AI models used to create content like marketing videos with AI-generated voices and avatars.

Example solutions: Jasper, Notion AI, Phrasee, HubSpot Content Assistant

5) Document Summarization

A simple but effective way to use generative AI in your business is to summarize long and complex documents.

This can be a quick and easy way to identify the most important points in a long contract, legal document, conversation transcript, or other long-form content type. While some generative AI tools are standalone content summarization tools, others are embedded in everyday business operations and project management tools.

For example, Notion AI is a feature that enables Notion users to quickly summarize project notes and sales call transcripts.

Example solutions: Cohere Summarize, Anthropic Claude, PatentPal, AI21 Studio

6) Project and Workflow Management

Several of the most popular project management platforms have or are currently adding generative AI assistive capabilities to their features lists. These assistants can help with administrative tasks like note-taking during meetings, email writing, and document summarization, freeing up time for your employees to focus on more strategic tasks.

At a more technical level, these AI assistants can also automate and optimize workflow builds, make task suggestions, and support data integration efforts.

Example solutions: Process Street, Wrike, Notion AI, ClickUp, Asana

7) Cybersecurity Management

Although generative AI can pose major risks to cybersecurity postures, these AI tools are also valuable for automating business cybersecurity management workflows and handling data security requirements.

Some of the most common ways businesses are taking advantage of generative AI in cybersecurity include through smarter threat intelligence and hunting, contextualized security environment recommendations, and attack simulations. Some of these new generative AI security tools are freestanding products, but many of them are either new features or add-on products for existing cybersecurity platforms.

Example solutions: Google Cloud Security AI Workbench, Microsoft Security Copilot, CrowdStrike Charlotte AI, Airgap Networks ThreatGPT

Also see: Top Generative AI Apps and Tools

5 Benefits of Leveraging Generative AI in Your Business

Depending on how and how much you choose to leverage AI capabilities in your organization, expect the following generative AI benefits from your efforts:

• Quicker delivery times on projects: Generative AI models can quickly generate responses, new content, useful data, and other things your organization may need at scale. This leads to more efficient production and allows your team to complete projects on tighter timelines, which is ultimately more cost-effective.
• Fewer hands-on employee requirements: Whether you lack the in-house staff expertise to complete a certain task or find that certain tasks are too tedious to hold employee attention, generative AI tools can pick up the slack and deliver consistent, accurate, and repeatable results.
• Support and QA for employee-driven tasks: Particularly with coding and product development tasks, but also with data analytics and other content forms, generative AI tools can act as quality assurance analysts, checking employee work for errors and sometimes correcting those errors without human intervention.
• Enhanced user experiences: Many generative AI tools are embedded in software that your employees or customers already use. With the help of generative AI, search engines, enterprise knowledge bases, project management platforms, and other user-facing tools are optimized for natural language inputs and an overall better user experience.
• Industry-specific solutions: Several generative AI models have been fine-tuned to address industry- and function-specific requirements across sectors like healthcare and pharmaceuticals, insurance, manufacturing, and more. If you’re operating in a specialized industry — even with strict compliance requirements — there’s more than likely a generative AI model that will address your needs and simplify your workflow.

Also see: Best Artificial Intelligence Software 2023

5 Challenges of Using Generative AI

Many risks come with using generative AI, and especially because this technology is still so new, not all ethical use dilemmas have been worked out. If you choose to use generative AI in your business, be on the lookout for these possible challenges you’ll face:

• Limited regulatory requirements and guidance: Generative AI is not currently regulated on a larger scale, and at this time, there are few comprehensive usage frameworks that cover how your organization can use generative AI compliantly and effectively.
• Lack of data and training transparency: Although many generative AI vendors are beginning to improve in this area, most are still not transparent with the types and sources of the training data they use, which leads to additional compliance and usage concerns.
• Cybersecurity concerns: In the wrong hands or in the hands of an ill-trained employee, sensitive data may be exposed to generative models and become part of the training set. Bad-faith actors can also train these models to work around their rules to complete unauthorized tasks.
• Identifying inaccurate outputs: Generative AI models use natural language so well that it’s sometimes difficult to determine if a model has generated an inaccurate response. Users will need to have the skills and nuanced knowledge to check their own work when using this type of technology.
• Employee usage errors: It’s difficult to regulate and control how employees use generative AI in their work. Coming up with an AI use policy is a good step toward mitigating these errors, but there’s still no consistent way to enforce usage rules, control what data employees use as inputs, or monitor multiple employees’ actions at once.

Also read: Generative AI’s Drawbacks: IP to Ethics

What Are Generative AI Best Practices for Business Usage?

Establishing best practices and procedures for generative AI use is the best way to manage internal business use of these models. Follow these best practices for better generative AI outcomes in your business:

• Use data from verified, credible, and approved sources as inputs.
• Develop an AI use policy and training plan that clearly outlines how and when employees can use generative AI tools.
• Partner with a generative AI vendor building models that can scale or be fine-tuned as your business requirements evolve; it’s also beneficial to identify vendors with products that integrate with your existing tool stack.
• Do your due diligence and research generative AI vendors’ policies, terms, and reputations before committing.
• Consider your budget and what tools fit into that budget; many foundation models are not that expensive but can quickly go up in price depending on your usage volume.
• Consider any generative AI models you use as part of your cybersecurity threat landscape and protect accordingly.

For more on this topic, read these comprehensive guides:

• Risks of Generative AI: 6 Risk Management Tips
• Generative AI Ethics: Concerns and Solutions
• Generative AI and Data Analytics: Best Practices

Bottom Line: Generative AI in Your Business

Generative AI technology is a powerful resource that can be leveraged in businesses of all sizes and backgrounds, especially since so many models come in affordable limited versions that still have extensive capabilities.

The most important thing to remember when using generative AI in your business is that these tools are only as effective as the users, inputs, and procedures that surround them. Make sure all employees are trained and given the resources they need to use generative AI in their work effectively, and you’ll achieve new levels of automation, smart assistance, and productivity in your organization.

Read next: AI Detector Tools

The post Generative AI for Business: 7 Productivity Boosts appeared first on eWEEK.

I spoke with Subha Tatavarti, CTO of Wipro, about the excitement, confusion and enormous potential of generative AI.

Among the topics we discussed: 

• As you survey the rapid rise of generative AI tools, what key trends are driving the market? What are the risks here?
• How can companies protect themselves from these risks even as they maximize their potential?
• How is Wipro addressing the generative AI needs of its clients? What’s the Wipro advantage?
• The future of generative AI tools? What do you see, say 1-3 years ahead?

Listen to the podcast:

Also available on Apple Podcasts

Watch the video:

The post Wipro CTO Subha Tatavarti on Generative AI appeared first on eWEEK.

In theory, the recent Google Next cloud computing conference was focused on cloud computing. As if to prove the point, Google pasted its marketing slogan all over the event in massive font: “the new way to cloud.”

In reality, I’ve never been to a cloud conference in which cloud played such a supporting role to another emerging technology – and I’ve attended numerous cloud shows.

So what was the other technology? You earn 5,000 bonus points if you can guess. Hint: high school seniors are using it to write their college entrance exams.

Okay, congrats – you guessed correctly. But given the trends in tech this year it should have been no surprise: clearly, generative AI was the true star of the Google cloud event.

This year some 8,000 attendees converged on the Moscone Center in San Francisco to talk shop, endure vendor pitches, and eat at some of the best (and most expensive) restaurants in the country. I’d be surprised if every last one of them didn’t at least mention AI as they mingled.

So what was everyone saying? To provide insight about where we are with artificial intelligence – and where we’re going – I spoke with six industry experts, a range of vendors and analysts.

For analysis of what the Google Cloud conference means for cloud and AI, and to read the expert predictions, go to:

• The (Real) Big News: The Future of Cloud and AI
• Gopal Srinivasan, Principal, Deloitte
  
• Sheenam Gupta, Broadcom
  
• Jennifer Marchand, Capgemini
  
• Ted Kwartler, DataRobot
  
• Ed Anuff, DataStax
  
• Dean Pratt, Kyndryl
  

The (Real) Big News: the Future of Cloud and AI

Google made more than 150 product announcements at the show – across data, cloud and artificial intelligence – and amid the deluge it was hard to know what mattered most.

As the flurry settled, it was clear: the biggest grand reveal was about Google’s enhanced partnership with Nvidia – in other words, the news that really counted was about AI.

The chipmaker Nvidia, of course, is at the very molten core of the AI frenzy gripping the tech world in 2023. Companies know they can’t obtain top AI performance without Nvidia’s GPUs. As the interest in AI has surged, Nvidia’s stock price has rocketed more than 200 percent this year; when Google announced the new alliance, Nvidia’s share price spiked 4.2 percent in one day.

The improved alliance means Google cloud customers can now more easily access Nvidia solutions. Google will host Nvidia’s DGX cloud, including the initial instances of DGX supercomputing, so Google customers will be able to deploy supercomputing capabilities. Additionally, by next month, Google will offer A3 virtual machines powered by Nvidia H100 GPUs. In other words, if you need access to Nvidia’s limited supply of ultra-fast AI processor chips, Google Cloud is a source.

But here’s where things get blurry. Sure, the news was big on the AI front, but it’s also cloud news. The news is a prime example of how cloud and AI are now joined at the hip.

The Google-Nvidia partnership reveals a major truth about AI’s future: AI can’t grow without cloud to support it. AI gets all the headlines these days, but in reality AI is part of a trilogy: artificial intelligence, cloud computing and data management all work together to support AI.

AI is a data hog; it requires vast oceans of data to do its work. And at the enterprise level, AI requires robust, scalable computing power – hence the need for a cloud platform like Google’s.

The fact that cloud shared the spotlight at the Google Cloud show – ostensibly focused on cloud – is not a negative about cloud’s fortunes. Just the opposite: the event’s deep focus on the AI-cloud relationship demonstrates the primacy of cloud.

In the big picture, this cloud-AI symbiosis means that, in the battle to control AI’s future, the large cloud providers are largely in control. That’s good news for Google, Amazon Web Services and Microsoft Azure. It’s challenging news for anyone seeking to compete with these mega players.

As for Google, the company was clearly happy to focus on AI at its annual cloud show. Among the big three in cloud, Google is a perennial third behind AWS and Azure. If the battle for market share is about cloud alone, it works against Google. The company is smart to play up its traditional strengths in data and AI.

However, the extent of Google Cloud’s benefit from its Nvidia alliance is unclear. Cloud customers can also access Nvidia’s solutions through AWS and Azure. And it’s probable that each cloud provider is nurturing its Nvidia partnerships on an ongoing basis.

The excitement in the tech world about Nvidia is frothy, to be sure. As I spoke with people at the Google conference, I heard numerous references to “Jensen” – the Nvidia CEO’s name is Jensen Huang. He has become such celebrity that people refer to him with one name, much like Drake or Beyonce.

Also see: 100+ Top AI Companies 2023

Vendors and Analysts: Expert Opinion

I spoke with the following experts at the Google Cloud event about the future of generative AI. Their comments have been edited for length and clarity.

Gopal Srinivasan

Principal, Deloitte, Generative AI Leader for Google Alphabet Alliance

I’ll start at the high level and then get into specifics over how AI might evolve. We’ve spent the last hundred years trying to abstract away how we as humans interact with computers. And we’ve gone from vacuum tubes to software and the internet – and generative AI is a new leap in how we interact with computers.

It’s going to become all-pervasive and be in every form of computing, from software applications, software in physical devices, and within the enterprise, across the board in every function, from finance to HR to customer service. Generative AI is going to become the way we build applications and deliver experiences for users.

How might this play out over 1, 3, 5 years? That obviously remains to be seen, but there’s already significant momentum, and over the next 12 months we’ll definitely see a category of applications that are centered on essentially assisting users. Meaning to help perform a workflow: help them perform faster, get them the insights and the information they’re looking for faster.

If you go beyond that in the second phase, you’ll see that much more augmentation. And the difference in going from assistance to augmentation is that AI will perform portions of the workflow entirely by itself.

So humans will perform certain pieces, AI will perform certain pieces end-to-end, and the entire workflow will get strung together.

Beyond that is the third phase, autonomy, where entire work processes within the enterprise will be performed end-to-end by AI. For example, your entire finance reconciliation process, end-to-end, can be done by AI with no human intervention. So broadly speaking, AI is going to essentially seep into every part of how we use technology within the enterprise and bring to bear the next form of how we use computing.

eWeek: Do you think that today’s executives fully understand the technology enough to fully plan for it?

Not quite, but what I would say is, reflecting on the past couple of evolutions, there is definitely a lot more interest and willingness to learn. I think most executives get the potential, they obviously have a lot of questions about the risks, and they’re more keen on understanding how the technology actually works.

So we are not there fully in terms of executives getting to the full potential, how it can transform their business. But that’s always [true] with new technology paradigms, the process of education and change management. And that’s happening more rapidly than I’ve ever seen in my 25 years of working in and around technology.

Sheenam Gupta

Lead Product Management, AIOps and Observability, Broadcom

The one thing that I’m seeing settling down now, as compared to when we talked about AI and ML five years back, is that there’s a realization [among executives] that AI/ML is not magic. It can provide outcomes if it is aligned to the right use cases or the right goals for an enterprise.

And so ideally what AI and ML needs to do is to augment knowledge. To augment [staffers] skills to help them avoid doing the mundane day-to-day tasks and leave it to the insights from AI or ML. Instead, have the teams focused more on innovation or adopting the newer technologies because that will also take a lot of experimenting, just like AI and ML did.

eWeek: What about the executives who have some doubts that AI can really play a leading role?

Oh yeah, we hear that all the time from the executives, they say, “are you telling me that because you have AI/ML in your product, it’ll replace all the tribal knowledge that my teams have? Or will it help us do some cost reductions?”

And our answer typically is a yes and a no. A no, because what AI and ML should essentially do is really augment what their teams have been doing. It should really learn from them and should support them. Or it should provide that next level for them to do things faster or to make sure their customers are getting the kind of experience they want to deliver with the help of ML. And a yes, because it would help their existing teams to channelize or to expand their scope of influence across [areas] more oriented with their business goals instead of staying stuck.

So our answer to them is that our machine learning solution will be a supervised model, which can take feedback from their employees; it should partner with them rather than competing and replacing them.

Also see: Top Generative AI Apps and Tools

Jennifer Marchand

Google Cloud COE Leader, the Americas, Capgemini

One interesting area we’ve started to build out with generative AI for our customers is conversational commerce, so really about expanding the channel reach to customers with minimal investment.

Take for example, a kiosk in a store concept. So you’re able to put a digital-human order kiosk in a partner store, and it’s ‘always on’ for business. Our focus group found that consumers are more willing to engage for longer with technology when you have the digital-human front end. And when you layer on the generative AI component to that, it creates much more dynamism with the consumer as well as the ability to upsell, cross-sell, and personalize that experience.

eWeek: What about generative AI’s effect on jobs and employment?  

I think that jobs will evolve and the people who will succeed are the people that embrace and learn how to use generative AI to make themselves more productive and competitive. But it’s all about finding and designing the use cases to keep the human in the loop.

Figure out the things that you don’t want to do manually. Give that to generative AI and add your value where it’s really going to have an impact.

Ted Kwartler

Field CTO, DataRobot

I think in the next one to three years, you’re going to see a merger of predictive AI – that “good old-fashioned AI” – with generative AI.

I think there’s going to be a merger of the two because we’ve seen that consumers want to interact with machine learning and AI in a way that’s easily consumed and contextualized for them. Bringing the two together makes a lot of sense.

In the interim, where we’re at right now, people have to make the choice of what model to use, and if they go off in the wrong direction, given the rate of change and the rate of innovation, they may end up building in a lot of tech debt.

So I think in the 12, 24 month timeframe, people that succeed will have to have a lot of flexibility so that they can avoid tech debt. I think in the end, we end up in a place that’s very valuable to the enterprise. I think we merge predictive and generative so that the end user has more context for what’s getting forecast.

Also see: Generative AI Companies: Top 12 Leaders

Ed Anuff

Chief Product Officer, DataStax

When we’re talking about the enterprise, it’s really interesting because the truism of the future being unevenly distributed is going to be valid in this case. We’ve got folks that are going to be doing all the stuff that people are talking about: chatbots and AI assistance live on their websites this year. But the majority of folks will be getting this stuff into production next year and the year after.

So the “crossing the chasm” still applies. However, if you’re not doing everything that everyone seems to be talking about, don’t worry about it. There’s still plenty of time to build the right things for your business and get them out there.

I do think costs are going to keep going down, and the power of the computation, whether it’s GPU or CPU-based, is going to keep going up.

The options for running these [large language] models are going to get more varied. You can now run Llama 2 on your laptop and people are doing it and getting great results. And if I can run Llama 2 on my laptop, it means I can run it in my data center as well. So we’ll see a lot of that.

We’ll see just massive new things in the next wave of these models. The stuff people are talking about in terms of their size and power – a lot of it is unpredictable. A lot of the stuff that we’re dealing with now is just going to get a lot more powerful and easier to apply.

But more power, lower cost is going to be the main thing. The tooling is going to get better. More of your developers are going to be able to use AI – not just your data scientists or your ML experts.

eWeek: What about the high cost of hiring AI experts?

Look, most of us who have been around for a while, we remember 10 years ago, 15 years ago, when every major brand, every enterprise, everyone in the Fortune 500 needed to build mobile apps and mobile developers were very expensive at that point. And many companies farmed out their mobile app development to consultants, digital agencies, and paid top dollar to do it because the opportunity was that pressing.

So I think the same thing is happening with AI right now. There’s inarguably a lot of skills that you need to make this happen: top-notch experts. But again, that’s a point in time. The tooling is getting better, so then the costs will go down.

A lot of people are surprised by that because there’s a lot of mystique around data science right now, and it’s improving very, very fast. If you look at a lot of these startups that are building these amazing apps on top of AI, these are people who are in their first exposure to AI. They had no previous experience prior to this year, yet you’re seeing them build these amazing things.

Dean Pratt

Google Cloud and Generative AI Evangelist, Kyndryl

Where I see generative AI headed, I think it will accomplish some of the technical skills that are needed today in a much faster way. We already have NLP to [create] code. We already have generative AI accomplishing a lot of the functions and summarizations and data gathering that we look for.

However, with all the new possibilities, there’s a wealth of pitfalls as well. New tools are extremely important, but understanding those in a responsible manner – being aware of the risk exposure when utilizing new tools as well as the beneficial outcomes – it’s important.

Where I’m excited with generative AI is in the diametric conversions of quantum computing and generative AI, and the expansive neural matrices we’ll see as a result – this should provide some very exciting outcomes.

We’re going to see the industry shift in monumental ways – I’m very excited to see that and also help lay the foundation for the responsible adoption of those tools.

Read next: Generative AI Startups

The post Experts Discuss AI at the Google Cloud Conference appeared first on eWEEK.

I spoke with Theresa Lanowitz, Head of Cybersecurity Evangelism at AT&T Business, about the issues the involved with securing an edge deployment.

Among the topics we discussed: 

• Let’s look at the edge market broadly. Clearly edge has seen rapid growth in the last few years. What trends are driving the edge market here in 2023? What are some key findings from the Edge Security Report?
• As companies plan and build out their edge deployments, what are the most challenging security concerns? What do you see companies struggling with the most?
• How do you recommend companies address these security challenges to their edge infrastructure?
• How is AT&T serving the edge computing needs of its clients?
• The future of edge computing and security, over the next 1-3 years? It seems like security in the edge sector has a long way to go.

Listen to the podcast:

Also available on Apple Podcasts

Watch the video:

The post AT&T’s Theresa Lanowitz on Cybersecurity in Edge Computing appeared first on eWEEK.

To provide a generative AI tool for business users, OpenAI has released ChatGPT Enterprise, an enterprise-grade version of the popular ChatGPT solution that expands on existing ChatGPT functionality.

With more consistent access to GPT-4 and a variety of new and enhanced security and collaboration features, ChatGPT Enterprise is an important step toward business adoption of the still new sector of generative AI.

Read on to learn more about what’s included in this latest release and how it could benefit your organizational processes and teams.

Table of Contents: ChatGPT Enterprise Release

• What Is ChatGPT Enterprise? Fast Facts
• New Features Available in ChatGPT Enterprise
• Benefits of Using ChatGPT Enterprise
• Bottom Line: What ChatGPT Enterprise Could Mean for Your Business

What Is ChatGPT Enterprise? Fast Facts

ChatGPT Enterprise already has several customers and advanced generative AI use cases available. Here are some of the highlights of what we know about the tool so far:

• Public release: ChatGPT Enterprise was released to the greater public on August 28, 2023.
• Early release users: Several enterprises were able to use ChatGPT Enterprise through early access, including Asana, Canva, PwC, Quizlet, Ramp, Zapier, and Khan Academy.
• Key features: ChatGPT Enterprise focuses on enterprise-grade security and privacy features, better GPT-4 usability and speed, and increased admin controls, among other features for enterprise-level operations and scalability.
• Cost: Unlike with its Free and Plus plans, OpenAI does not transparently list pricing information for ChatGPT Enterprise. Interested users will need to contact the OpenAI sales team directly for pricing.
• Future enterprise AI roadmap: OpenAI has revealed some of the Enterprise AI features they will release next, including additional application-level customizations, more robust power tools for data analysis, and role-specific features and functions.

Also see: Top Generative AI Apps and Tools

New Features in ChatGPT Enterprise

While certain aspects of ChatGPT Enterprise look like its predecessors, this latest release adds to and expands upon the generative AI model’s security, admin, and data analytics features. Take a closer look at the new features in ChatGPT Enterprise:

Security, Privacy, and Compliance

The main area that OpenAI focuses on in its new release announcement is security and privacy for generative AI. The new enterprise model includes both data encryption at rest and in transit through AES 256 and TLS 1.2+, SOC 2 compliance features and certification, and SSO and domain verification features.

But more significantly, OpenAI has promised that this model will not use customer prompts and queries or any other customer data to train future iterations of the model.

This means organizations can confidently input their data for various business use cases without fear of exposing proprietary business data. This new security and privacy commitment from OpenAI has opened ChatGPT up to new users and industries, especially those that require built-in compliance features and security safeguards to protect either consumer or top-secret business data.

Team Management and Admin Features

ChatGPT Enterprise addresses enterprise needs by giving internal administrators and teams more features and functions. For example, the new admin console gives admins access to a dashboard with a database for bulk member management, easy-to-read analytics and charts for usage insights, configurable admin settings, and identity and provisioning features and actions.

At a team level, shareable chat templates are now available and make it possible for teams to build collaborative, repeatable, and standardized projects and workflows. These features are designed to support a smoother user experience across an enterprise environment as well as easier large-scale deployments for business leaders and IT teams.

Advanced Data Analytics

Advanced data analysis for generative AI are available to ChatGPT Enterprise users on an unlimited basis. These capabilities, previously under the label Code Interpreter, are designed to support more technical data analysis tasks as well as democratized, citizen data scientist projects across product design and testing, financial analysis, ETL, and other analytics and data management use cases.

Speed and Scalability Features

This latest version of ChatGPT has increased its speed and power to meet the requirements of enterprise-level operations. Users of ChatGPT Enterprise receive access to unlimited GPT-4, longer context windows, and higher speeds (depending on system utilization).

OpenAI has built up this version of ChatGPT with 32k context for “4x longer inputs, files, or follow-ups,” which makes it possible for users to manage longer conversational strings and problem-solving scenarios.

Customization Features

ChatGPT Enterprise is one of the most customizable products from OpenAI to date, allowing users to do many tasks through self-service features.

Prebuilt chat templates can be used to simplify and standardize team-wide workflows. And for teams that need customizations that are more specific to their industry or business use case, ChatGPT Enterprise comes with several API credits at no additional cost.

Also see: Best Artificial Intelligence Software 2023

Benefits of ChatGPT Enterprise

ChatGPT Enterprise offers a variety of benefits that previous iterations of the tool did not, including the following:

• Larger and more effective context windows: Because this version of ChatGPT can process four times more input and file content at a time than previous versions of the tool, it can effectively generate useful and contextualized responses based on greater quantities of older input data.
• New customizability opportunities: The admin dashboard and many other features can be customized without much effort. Enterprise users also have access to free API credits if they require additional customizations.
• Better privacy and compliance features and settings: ChatGPT Enterprise has new and enhanced encryption features in place as well as SOC 2 compliance. These features, along with the company’s updated privacy commitments and the tool’s APIs for customizability, make it possible for organizations to rightsize ChatGPT to their privacy and compliance requirements.
• Business data protection: With this model, OpenAI has promised to prevent its model from ingesting and using user inputs for future model training and dataset incorporation.
• Democratized data analysis assistance: Users get unlimited access to OpenAI’s advanced data analysis tools, which are designed to support both traditional data science and low-code/no-code democratized data science initiatives.
• Increased team productivity: Current users have seen their organization’s productivity increase with the help of this tool. According to Jorge Zuniga, the head of data systems and integrations at Asana, “ChatGPT Enterprise has cut down research time by an average of an hour per day, increasing productivity for people on our team. It’s been a powerful tool that has accelerated testing hypotheses and improving our internal systems.”
• Unlimited GPT-4 access: This is the first model that has unlimited access to GPT-4’s higher speeds and greater processing capacity.
• Promising future roadmap: OpenAI has already publicly announced some of its future plans, which look promising for organizations that want more self-service features, stronger data analysis capabilities, function-specific features, and more customizability.

Also see: Generative AI Companies: Top 12 Leaders 

Bottom Line: What ChatGPT Enterprise Could Mean for Your Business

ChatGPT Enterprise has now been publicly released, and OpenAI says it is “onboarding as many enterprises as [they] can over the next few weeks.” Several enterprises have already committed to and praised the tool, explaining how well it addresses their more specific generative Ai enterprise use cases with added security and privacy features.

While the current version of this tool may not fit into the budget or address the needs of a smaller business, OpenAI’s future roadmap indicates that this is only the beginning of self-service-driven, customizable models for them. It will be interesting to see how other AI vendors respond and how the generative AI landscape changes over the next several months to further meet enterprise AI expectations.

Read Next: Top Natural Language Processing Companies

The post ChatGPT Enterprise: AI for Business appeared first on eWEEK.

By leveraging natural language processing and generative AI, conversational AI platforms enable businesses to build intelligent AI chatbots and virtual assistants that can understand and respond to user queries seamlessly.

We highlight the top Conversational AI platforms empowering enterprises to deliver personalized, efficient, and engaging customer experiences.

Here are our picks for the best conversational AI platforms:

• IBM Watson Assistant: Best for advanced features
• Kore.ai: Best for automation
• Avaamo.ai: Best for conversational analytics
• Amazon Lex: Best for affordability
• Oracle Digital Assistant: Best for performing operational tasks
• Microsoft Bot Framework: Best for developers
• Cognigy.AI: Best for contact center automation
• Yellow.ai: Best for customization
• Acquire: Best for ease of use and simplicity
• Clinc: Best for financial service companies
• Top conversational AI platform: Comparison chart
• Top conversational AI platform for your business
• Key features of conversational AI platforms
• Benefits of working with conversational AI platforms
• How do I choose the best conversational software for my business?
• How we evaluated conversational software
• Frequently Asked Questions (FAQ)
• Bottom Line: Conversational AI Platforms

Top Conversational AI Platform: Comparison chart

Here is a head-to-head comparison summary of the best conversational AI platforms.

IBM Watson Assistant: Best for advanced features

Built on deep learning, machine learning, and natural language processing (NLP) models, the IBM Watson Assistant conversational AI platform enables your teams to build AI-powered voice agents and chatbots that deliver automated self-service support across all channels and touchpoints. It can search and find answers to customer inquiries in existing documents, websites, and knowledge bases in order to complete the user’s intended action.

The platform allows you to build an AI chatbot that can be trained to understand user requests and adapted to your business scenarios – it also can recognize plain-language responses from your customers, like synonyms, dates, times, and numbers. It integrates with various third-party services, including WhatsApp, Slack, Facebook Messenger, Kustomer, Zendesk and more.

Pricing

IBM Watson is available for free with basic features and paid versions with advanced features.

• Lite: Available at no cost. It includes 1,000 unique monthly active users (MAUs), up to 10,000 messages per month, up to five skills (Dialog, Action, Search), seven days of usage analytics, session inactivity timeout of 5 minutes, and services deleted after 30 days of inactivity.
• Plus: Starts at $140 per month. It supports phone and SMS integration and includes 1,000 MAUs. Additional MAUs are billed at $14 per 100 MAUs.
• Enterprise: Quotes available on request.

Features

• Supported channels include webchat, telephony and voice add-ons, SMS and MMS, messaging channel integrations and custom channel API.
• Customer service desk integrations.
• Supports up to 100 concurrent call capacity for its plus plan and up to 1,000 for its enterprise plan.
• Deploy on any cloud, including IBM, Amazon, Google, Microsoft, or on-premises environments.

Pros

• Natural Language Processing (NLP) capability.
• Multi-language support.
• Language, acoustic and voice customization options.

Cons

• Limited duration for its analytics data retention capability.
• Onboard support is only available to enterprise users.

Also see: Best Artificial Intelligence Software 2023

Kore.ai Experience Optimization (XO) Platform: Best for automation

Kore.ai Experience Optimization (XO) Platform is designed to help businesses build, train, and deploy intelligent chatbots and virtual assistants that can understand and respond to customer queries in a natural, conversational manner.

Aside from the XO platform, the company also offers various solutions such as SmartAssist, an AI-native CCaaS solution; AgentAssist, which are virtual assistants supporting agents in real time; and BankAssist, which is an AI-native, omnichannel, conversational banking assistant pre-trained on 400+ retail banking use cases.

Pricing

• Standard: 20 cents per conversation. This plan also includes 10 cents per conversation for pre-built apps and 4 cents per minute for voice automation.
• Enterprise: Custom quotes.

Korea.ai offers optional enhanced support at an additional cost – $2,000 per month for the standard plan and a custom quote for the enterprise plan.

Features

• Support over 100 languages, including English, Arabic, Japanese, French, and Malay.
• Auto-dialog generation.
• On-premise, cloud and hybrid deployment options.
• Conversational IVR (Interactive Voice Response) capability.

Pros

• Integrates with over 75 prebuilt workflows, including CRM, ERP, ITSM, and NLM tools.
• Compliant with industry regulations like SOC2 Type2, HIPAA, GDPR and PCI.
• Easy to use.

Cons

• Advanced support costs extra.
• The documentation can be improved.

Avaamo.ai: Best for conversational analytics

Powered by neural networks, speech synthesis, and deep learning, Avaamo is a conversational artificial intelligence platform that provides businesses with intelligent virtual assistants and chatbots. Avaamo offers fabricated skillsets to help enterprises automate complex business use cases through multi-turn conversations.

Avaamo offers a skills builder that includes a flow designer for designing conversation, dynamic dialog, conversational IVR, and other tools that enable you to automate complex enterprise use cases.

Pricing

Avaamo doesn’t advertise pricing on its website; the company encourages users to request a demo to learn about the platform and get custom quotes based on their needs.

Features

• It supports integration using APIs, Web Services, ESBs, MQs, and Custom Adapters, including over 150 pre-built integrations.
• Branded AI voice capability.
• Conversational analytic tools.
• Compliance with industry regulations such as SOC3, SOC2(Type Ⅱ), CCPA, HIPAA and more.
• IVR channel integration.

Pros

• Supports over 114 languages.
• Flexible deployments – on-premise and cloud deployment.
• It supports over 70 channels across various chats and voice mediums.

Cons

• Some users reported that Avaamo is somewhat expensive.

Amazon Lex: Best for affordability

Amazon Lex is a chatbot-building technology by AWS, so it benefits from the cloud giant’s extensive platform. It’s a service that can help you build conversational interfaces using voice and text. Amazon Lex uses Automatic Speech Recognition (ASR) to convert spoken language into text and Text-to-Speech (TTS) to convert text into spoken language. The platform has various components that work together to enable users to build intelligent conversational AI solutions for their businesses. They include:

• Intent: It represents the desired action.
• Utterance: Phrases or sentences that users input to interact with the chatbot.
• Prompt: Request data.
• Slot: Required data.
• Fulfillment: Amazon Lex can trigger backend business logic to fulfill user requests.

Pricing

Amazon Lex only charges you for what you use. You can also try the tool for free. You can process up to 10,000 text requests and 5,000 speech requests or speech intervals per month at no cost for the first year of trying the tool.

Request and response interaction pricing

• $0.004 per speech request.
• $0.00075 per text request.

Streaming conversation pricing

• $0.0065 per unit per speech interval.
• $0.0020 per unit per text request.

Automated chatbot designer pricing

• $0.50 charges per minute

You can use the AWS pricing calculator to calculate your monthly expense or contact an AWS specialist for pricing assistance to get an estimate.

Features

• Natural conversation capability includes speech recognition and natural language understanding, context management, 8 kHz telephony audio support and multi-turn dialog.
• Builder productivity capability that allows you to build visual conversations, streamline conversations and deploy to multiple platforms.
• Integration with AWS services like Amazon Kendra, Amazon Polly and AWS Lambda.
• Contact center integrations with Amazon Connect, Genesys Cloud CX, Amazon Chime SDK and AWS Contact Center Intelligence (CCI).

Pros

• Offers use cases for financial services, insurance, retail, telecom, and travel.
• Easy to train and configure chatbots.
• Positive reviews from current and past users.
• Users find the solution highly affordable.
• Natural language understanding.

Cons

• Documentation can be improved.
• Some users say the confidence score isn’t always accurate.

Also see: Top Generative AI Apps and Tools

Oracle Digital Assistant: Best for performing operational tasks

With the Oracle Conversational AI platform, you can build chatbots that can engage in natural language conversations, understand user intents, and provide relevant responses and actions. The platform lets you connect with a chatbot through channels like Microsoft Teams or Facebook on your website or embedded inside your mobile app.

Pricing

Oracle Digital Assistant platform pricing is available per request and as a subscription for SaaS customers by employee, named users, or sessions.

• Oracle Digital Assistant Cloud Service: $0.0232 unit price per request. This plan requires a minimum of 250 requests per hour.
• Oracle Digital Assistant Platform for Oracle SaaS – Hosted Employees: $3 per month.
• Oracle Digital Assistant Platform for Oracle SaaS – Hosted Named User: $6 per month.
• Oracle Digital Assistant Platform for SaaS – 1,000 Sessions: $275 per month.

Features

• NLU’s capability to understand conversations to derive intent and context.
• AI-powered voice.
• Prebuilt chatbots for Oracle Cloud applications such as Oracle Cloud ERP, Oracle Cloud SCM, Oracle Cloud HCM, and Oracle Cloud CX.

Pros

• Multilingual support.
• Multi-channel support: web, mobile, SMS texting, smart speakers, Slack, and Microsoft Teams.
• Users applaud its seamless integration with messaging apps.

Cons

• Limited language support.
• Voice recognition can be improved.

Microsoft Bot Framework: Best for developers

Supporting both code-first and no-code approaches, Microsoft Bot Framework provides a platform for creating conversational agents capable of interacting with users via channels such as online sites, Slack, and Facebook. The framework comprises various components:

• A language component that enables you to build natural language experiences.
• Speech component that can reply to customers in a natural language with a branded voice, hear commands, translate, and identify individual speakers.
• QnA Maker, which is a question-and-answer bot.
• Vision components that leverage the capabilities of computer vision services to recognize faces, moderate content, and index images and video.
• Search functionality allows you to search across defined domains or the web.

Pricing

• Standard channels: Free for unlimited messages.
• Premium channels: Free for up to 10,000 monthly messages or $0.50 per 1,000 for the S1 tier.

Features

• It can integrate into existing ecosystems such as Slack, Cortana, Skype, Telegram, Messenger, and SMS.
• Native integration with Azure cognitive services.
• Bot Builder SDK.

Pros

• Offers open source SDK and tools to connect your bot to popular channels and devices.
• Supports integration with Microsoft Azure Language Understanding (LUIS) for entity recognition and intent understanding.
• Rich development ecosystem.

Cons

• Limited customization options.
• Steep learning curve for beginners.

Also see: Generative AI Companies: Top 12 Leaders 

Cognigy.AI: Best for contact center automation

Enhanced with generative AI, Cognigy’s low code Conversational AI platform enables enterprises to automate contact centers for customer and employee communications. The platform offers customer service solutions like Conversational IVR, Smart Self-Service, and Agent + Assist. With Cognigy, users can design conversational flows, integrate with backend systems, and customize the behavior of their chatbots or virtual assistants to suit their specific business needs.

Pricing

The platform doesn’t advertise its pricing on its website. To get quotes, businesses are required to contact the company for a demo to discuss their needs. Publicly available information revealed the platform costs as follows.

• Basic 5K pm: Platform & Setup fee, 60K conversations pa, Standard Support costs $43,080 for 12 months.
• Basic 5K pm + VG: Platform & Setup fee, 60K conversations pa, 5 VGs, Standard Support costs $53,916 for 12 months.

Features

• Generative AI capability.
• AI-based voice gateway services.
• Omni-Channel Support – deployment across various channels, including websites, messaging apps, and voice assistants.
• Integrates with messaging apps like Teams, Slack, andWhatsApp; CCaaS and CPaaS such as Avaya, 8X8, NICE, and GENESYS; enterprise tools like Salesforce, ServiceNow, SAP, Zendesk, and Microsoft Dynamic 365.

Pros

• Reporting and analytics capabilities.
• Real time insight into your operations.

Cons

• Some users reported difficulty with building custom extensions.
• Limited documentation.

Yellow.ai: Best for customization

Our analysis found that Yellow.ai is a battle-tested conversational AI platform used by over 1,000 enterprises across 70 countries. Yellow.ai dynamic automation platform is designed to automate customer and employee interaction and conversations across text, email, and voice. The platform is fluent in over 135 languages and supports over 35 channels.

Pricing

Yellow.ai AI chatbot pricing are as follows:

• Free: Available at no cost for 100 MTU per month and limited to two agents.
• Enterprise: Usage-based pricing. Quotes are available upon request.

Publicly available information on the AWS marketplace reveals that the Yellow.ai Basic Plan, which includes a basic chatbot with one use case and tiered conversation limits, costs $10,000 for 12 months, while the Standard Plan, which includes a Standard chatbot with four use cases and tiered conversation limits, costs $25,000 for 12 months.

Features

• No code builder.
• Integrates with 100+ third-party services, including Zendesk, Freshdesk, Jira, ServiceNow, HubSpot, and Jira.
• Supports up to 135 languages.
• Text automation capability.
• Supports 35+ channels, including Slack, Messenger, Instagram, and WhatsApp.
• Pros
• Natural language understanding capability.
• Highly customizable.
• Drag and drop studio for building bots.

Cons

• Advanced features may be complex to implement.
• Some users reported that the integration of Yellow.ai with other tools is somewhat complex.

Acquire.io: Best for ease of use and simplicity

Featuring live chat, video and voice calling, AI chatbots, co-browsing and centralized interaction management, Acquire conversational AI platform empowers users to help customers resolve complex issues in real time. The platform aims to improve customer satisfaction, increase conversions, and enhance customer support efficiency.

Pricing

• Self service: This plan costs $500 per month billed annually plus $25 per agent per month.
• Integration solution: This plan costs $2,000 per month billed annually plus  $45 per agent per month.

Features

• One-way and two-way video capability.
• Live chat messaging.
• Translate conversations into 25+ languages.
• Dynamic prioritization and intelligent routing.

Pros

• Voice and video call options.
• Call deflection capability.
• Allows users to manage multiple chats simultaneously.

Cons

• Some users reported that the notification capability doesn’t function well, causing them to miss live chats.
• It may be somewhat expensive for small businesses.

Clinc: Best for financial service companies

Clinc’s AI platform is designed to provide personalized and natural language-based experiences for applications like virtual assistants, chatbots, and voice-controlled devices. The company has been particularly involved in developing  AI systems for the financial and banking sectors, where their technology is used to create virtual banking assistants that allow customers to check balances, report lost or stolen cards, transfer funds, and gain deep insights into their spending habits.

Pricing

Pricing is available upon request.

Features

• Flexible deployment – cloud or on-premise.
• IVR capability.
• Advanced Natural Language Processing.
• Omni-channel support – Alexa, web messenger and mobile apps.

Pros

• Voice and text support.
• Multi-lingual support.
• Context retention capability.

Cons

• Initial setup complexity.
• Documentation can be improved.

Key Features of Conversational AI Platforms

Natural Language Processing

The computer’s ability to understand human spoken or written language is known as natural language processing. NLP combines computational linguistics, machine learning, and deep learning models to process human language. This feature enables the conversational AI system to comprehend and interpret the nuances of human language, including context, intent, entities, and sentiment.

Omini-channel experience

The best conversational AI platform allows you to engage with your customers across multiple channels and touchpoints such as websites, mobile apps, messaging platforms (like Facebook Messenger, Instagram or WhatsApp), voice assistants (such as Amazon Alexa or Google Assistant), and more. The tools we analyzed in this guide allow you to engage with your customers across various channels.

Sentiment Analysis

The best conversational AI tools are trained to analyze digital text to deduce the emotional tone of the message – which could be positive, negative, or neutral. This capability allows chatbots to respond to customers in a more personalized way or empathetic manner.

Integration with third-party services

Integration capability is an important feature of any modern-day digital solution, especially for conversational AI platforms. Seamless integration with third-party services like CRM systems, messaging platforms, payment gateways, or ticketing systems allows businesses to provide personalized experiences.

Analytics and insights

Conversational AI platforms often provide analytics and insights into user interactions. This data can help businesses understand user behavior, identify common queries, and improve the effectiveness of the AI system.

How to Choose the Best Conversational AI Software for Your Business

While there are hundreds of conversational AI platforms marketed as the best system, each tool has its strengths and areas of weaknesses. What seems to be a strength for company A may be a weakness for company B and vice versa. Choosing the right conversational AI software for your business can significantly impact customer interactions, operational efficiency, and overall success. So, how do you select the best conversational AI platform for your business?

• Identify your conversational AI needs – the current gap in your existing system, and also define your expectations.
• Research various platforms and narrow your list to the best three that stand out.
• Evaluate these tools’ features and compare them to your requirements list.
• Select the platform with features that align with your needs.
• Consider the affordability of the tool, ease of use/usability and scalability.
• Research the vendor reputation – a well known, established vendor is more likely to provide reliable software and support. Then again, there are eager start-ups that provide a lot of value for the money.

Keep in mind that the best conversational AI software for your business will depend on your unique needs, goals, and the preferences of your customers.

How We Evaluated the Best Conversational AI Platforms

The best conversational AI platform is easy to use, offers features that meet the intended users’ needs, balances quality service and affordability, and allows businesses to integrate with tools and services they already use.

Core features – 30%

We evaluated each platform’s core offerings and their ability to serve the needs of businesses in various industries. Our analysis considered features like NLU, multi-channel support, flexible deployment, multi-lingual and other essential features.

Cost – 25%

We researched each conversational AI platform pricing strategy and model. We also checked for pricing transparency and the availability of free demos and trials to allow potential buyers to test out the platform before making a purchase decision.

Integration – 15%

We checked whether the conversational AI platform integrates with third party services such as CRM, ITSM, and various communication channels such as websites, messaging apps, voice assistants, and social media platforms.

Ease of use – 15%

We reviewed each platform’s usability to determine if it’s beginner-friendly. A user-friendly dashboard makes it easier for non-technical team members to manage the AI. So we checked if the platform has an intuitive interface for setting up and managing conversational flows.

Customer support – 15%

To score each conversational AI platform for this category, we analyzed user feedback on review sites and considered the types of support offered by each company.

Frequently Asked Questions (FAQs)

What features should I look for when evaluating conversational AI platforms?

Consider the following features when shopping for a conversational AI platform:

• Natural Language Understanding (NLU)
• Multi-turn dialog management
• Intent recognition
• Entity recognition
• Integration with existing systems
• Analytics and reporting
• Multi-channel support
• Language support
• Testing and debugging tools

What are conversational AI platforms and how do they work?

Conversational AI platforms are software solutions that leverage the innovations of AI, deep learning, and NLP technologies to enable automated, human-like interactions between computers and users through natural language.

These platforms facilitate seamless communication across various channels, such as chatbots, voice assistants, messaging apps, websites, and more. They are designed to understand user inputs, interpret their intentions, and provide relevant and contextual responses.

Which industries can benefit from adopting conversational AI platforms?

The potential applications of conversational AI are vast. Any industry that involves customer interactions, information dissemination, and process automation can benefit from leveraging conversational AI platforms.

Bottom Line: Conversational AI Platforms

As businesses seek to enhance their customer interactions and improve overall efficiency, conversational AI platforms offer organizations the opportunity to forge stronger connections with their audience, provide instant assistance, and gather valuable insights from user interactions. From understanding user intent to generating coherent responses, conversational AI platforms help business create lifelike conversations that meet customer needs efficiently.

Read Next: Top Natural Language Processing Companies

The post Top 10 Conversational AI Platforms 2023 appeared first on eWEEK.