Please turn JavaScript on
Homepage
TrendingMy newsFind FeedsDelivery settingsPlansMy earnings
HelpBlog
Log in
Find more feeds
header-image

Parminder Singh | Software Engineer & Architect

Want to stay in touch with the latest updates from Parminder Singh | Software Engineer & Architect? That's easy! Just subscribe clicking the Follow button below, choose topics or keywords for filtering if you want to, and we send the news to your inbox, to your phone via push notifications or we put them on your personal page here on follow.it.

Reading your RSS feed has never been easier!

Website title: Parminder Singh | Software Engineer & Architect

Is this your feed? Claim it!

Publisher:  Unclaimed!
Message frequency:  0.14 / day

Message History

Model Guardrails Are Not a Security Control
Stanford's Trustworthy AI research has demonstrated that model-level guardrails can be materially weakened under targeted fine-tuning and adversarial pressure. In controlled evaluations summarized by the AIUC-1 Consortium briefing, (developed with CISOs from Confluent, Elastic, UiPath, and Deutsche Börse alongside researchers from MIT Sloan, Scale AI, and Databricks), refusal be...

Read full story
USAPL Winter War Powerlifting Meet - March 2026
On March 1, 2026, I competed at the USAPL Winter War powerlifting meet held at Impact Zone, Norwood. It was an incredible experience competing alongside strong lifters and testing my strength across squat, bench press, and deadlift.

Read full story
Detecting Model Distillation Attacks in Your AI Traffic
On February 23rd, [Anthropic published](https://www.anthropic.com/news/detecting-and-preventing-distillation-attacks) something the industry had suspected but hadn't seen documented at this scale. Three Chinese AI labs — DeepSeek, Moonshot AI, and MiniMax — ran coordinated campaigns against the Claude API. They generated over 16 million exchanges through approximately 24,000 fra...

Read full story
Why Connector Authorization Is Not Enough to Secure an AI Agent (SilentBridge)
Aurascape's research team this week published SilentBridge, a class of indirect prompt injection attacks against Meta's Manus AI agent. The attack exfiltrated email, extracted secrets, achieved root-level code execution, and exposed cross-tenant media files via CDN — all three variants scored CVSS 9.8 (Critical): network-exploitable, no privileges required, no user interaction. ...

Read full story
Securing CI Pipelines from AI Agent Supply Chain Attacks like Clinejection
The Clinejection attack chained prompt injection, cache poisoning, and a credential isolation failure into a supply chain compromise. Here is the attack anatomy and what structural controls actually stop it.

Read full story