AI Impacts

Energy efficiency of monarch butterfly flight

Thu, 26 Nov 2020 06:30:47 GMT

Updated Nov 25, 2020

According to very rough estimates, the monarch butterfly:

can fly around 100,000-600,000 m/kJ
and move mass at around 0.065-0.36 kg.m/J

Details

The Monarch Butterfly is a butterfly known for its impressively long migration.¹

Mass

The average mass of a monarch butterfly prior to its annual migration has been estimated to be 600mg²

Distance per Joule

The following table gives some very rough estimates of energy expenditures, speeds and distances for several modes of flight, based on confusing information from a small number of papers (see footnotes for details).

Activity	Description	Energy expenditure per mass ( J/g.h)	Energy expenditure for 600mg butterfly (J/s)	Speed (m/s)	distance/energy (m/J)
Soaring/gliding	Unpowered flight, including gradual decline and ascent via air currents	8-33³	~0.0014 – 0.0056⁴	Very roughly 2.5-3.6 on average⁵	446- 2571⁶
Cruising	Low speed powered flight	Very roughly 209⁷	0.035⁸	Maximum: >5⁹	Maximum: >143¹⁰
Sustained flapping	High speed powered flight	Very roughly 837¹¹	~0.14¹²	Maximum: >13.9¹³	Maximum: >99¹⁴

Table 1: Statistics for several modes of flight. All figures are very rough estimates, based on incomplete and confusing information from a small number of papers (see footnotes for details).

Soaring is estimated to be potentially very energy efficient (see Table 1), since it mostly makes use of air currents for energy. It seems likely that at least a small amount of powered flight is needed for getting into the air, however monarch butterflies can apparently fly for hundreds of kilometers in a day¹⁵, so supposing that they don’t stop many times in a day, taking off seems likely a negligible part of the flight.¹⁶

This would require ideal wind conditions, and our impression is that in practice, butterflies do not often fly very long distances without using at least a small amount of powered flight.¹⁷

There is stronger evidence that monarch butterflies can realistically soar around 85% of the time, from Gibo & Pallett, who report their observations of butterflies under relatively good conditions.¹⁸ So as a high estimate, we use this fraction of the time for soaring, and suppose that the remaining time is the relatively energy-efficient cruising, and take the optimistic end of all ranges. This gives us:

One second of flight = 0.15 seconds cruising + 0.85 seconds soaring

= 0.15s * 5 m/s cruising + 0.85s * 3.6m/s soaring

= 0.75m cruising + 3.06m soaring = 3.81m total

= 0.75m / 143 m/J cruising + 3.06m / 2571 m/J soaring = 0.0064 J total

Thus distance/energy = 3.81m/0.0064 J = 595 m/J

For a low estimate of efficiency, we will assume that all of the powered flight is the most energetic flight, that powered flight is required half the time on average, and that the energy cost of gliding is twice that of resting. This gives us:

Energy efficiency = (50% * soaring distance + 50% * powered distance)/(50% * soaring energy + 50% * powered energy)

= (50% * soaring distance/time + 50% * powered distance/time)/(50% * soaring energy/time + 50% * powered energy/time)

= (0.5 * 2.5m/s + 0.5 * 13.9m/s)/(0.5 * (0.0056 * 2) J/s + 0.5 * 0.14 J/s)

= 108 m/J

Thus we have, very roughly:

distance/energy = 100,000-600,000 m/kJ

Mass.distance per Joule

As noted earlier, the average mass of a monarch butterfly prior to its annual migration has been estimated to be 600mg¹⁹

Thus we have:

mass*distance/energy = 0.0006 kg * 108 — 0.0006 kg * 595 m/J

= 0.065 — 0.36 kg.m/J

Notes

Energy efficiency of wandering albatross flight

Wed, 25 Nov 2020 00:13:13 GMT

Updated Nov 24, 2020

The wandering albatross:

can fly at around 481m/kJ
and moved mass at around 2.8—6.1 kg.m/J

Details

The wandering albatross is a very large seabird that flies long distances on wings with the largest span of any bird.¹

Distance per Joule

Speed

In a study of wandering albatrosses flying in various wind speeds and directions, the fastest ground speed measured appears to be around 24m/s, ²

Energy expenditure

One study produced an estimate that when flying, albatrosses use 2.35 times their basal metabolic rate³ which same paper implies is around 1,833 kJ/bird.day.⁴

That gives us a flight cost for flying of 0.050 kJ/second.⁵

Distance per Joule calculation

This gives us a distance per energy score of:

distance/energy

= 24 m/s / 0.050 kJ/s

= 481m/kJ

Mass.distance per Joule

Albatrosses weigh 5.9 to 12.7 kg.⁶

Thus we can estimate:

mass.distance/Joule

= 5.9kg * 0.48 m/J to 12.7kg*0.48 m/J

= 2.8—6.1kg.m/J

Notes

Energy efficiency of paramotors

Tue, 24 Nov 2020 21:11:22 GMT

Updated Nov 24, 2020

We estimate that a record-breaking two-person paramotor:

covered around = 0.36 m/kJ
and moved mass at around 0.058 – 0.10 kg.m/J

Details

Paramotors are powered parachutes that allow the operator to steer.¹

Distance per Joule

The Fédération Aéronautique Internationale (FAI) maintains records for a number of classes of paramotor contest. We look at subclass RPF2T—(Paramotors : Paraglider Control / Foot-launched / Flown with two persons / Thermal Engine)—which is appears to be the most recent paramotor record for ‘Distance in a straight line with limited fuel’.²

The record distance was 123.18 km.³ The FAI rules state that no more than 7.5 kg of fuel may be used.⁴ We will assume that in the process of breaking this record, all of the available fuel was used. We will also assume that regular gasoline was used. Gasoline has an energy density of 45 MJ/kg.⁵

Distance per energy = 123.18 km / (7.5 kg * 45 MJ/kg)

= 0.36 m/kJ

Mass.distance per Joule

The weight of an entire paramotoring apparatus appears to be the weights of the passengers plus motor plus wing plus clothing and incidentals, based on forum posts.⁶ These posts put clothing and incidentals at around 8kg, but are estimates for single person flying, whereas this record was a two person flight. We guess that two people need around 1.5x as much additional weight, for 12kg.

Wikipedia says that the weight of a paramotor varies from 18kg to 34 kg.⁷ However it is unclear whether this means the motor itself, or all of the equipment involved.

The glider used appears to be MagMax brand, a typical example of which weighs around 8kg, though this may have been different in 2013, or they may have used a different specific glider.⁸ To account for this uncertainty, we shall add the glider weight to the high estimate, and so estimate the weight of the glider and motor together at 18-42kg.

We will assume that the apparently male pilots weighed between 65 and 115 kgs each, based on normal male weights⁹.

Thus we have:

weight = motor + wing + people + clothing and incidentals

weight (low estimate) = 18 + 65*2 + 12 = 160kg

weight (high estimate) = 42 + 115*2 + 12 = 284kg

High efficiency estimate:

284kg * 0.36 m/kJ = 0.10 kg.m/J

Low efficiency estimate:

160kg * 0.36 m/kJ = .058 kg.m/J

This gives us a range of 0.058 – 0.10 kg.m/J

Notes

Misalignment and misuse: whose values are manifest?

Thu, 19 Nov 2020 00:06:34 GMT

By Katja Grace, Nov 18 2020, Crossposted from world spirit sock puppet.

AI related disasters are often categorized as involving misaligned AI, or misuse, or accident. Where:

misuse means the bad outcomes were wanted by the people involved,
misalignment means the bad outcomes were wanted by AI (and not by its human creators), and
accident means that the bad outcomes were not wanted by those in power but happened anyway due to error.

In thinking about specific scenarios, these concepts seem less helpful.

I think a likely scenario leading to bad outcomes is that AI can be made which gives a set of people things they want, at the expense of future or distant resources that the relevant people do not care about or do not own.

For example, consider autonomous business strategizing AI systems that are profitable additions to many companies, but in the long run accrue resources and influence and really just want certain businesses to nominally succeed, resulting in a worthless future. Suppose Bob is considering whether to get a business strategizing AI for his business. It will make the difference between his business thriving and struggling, which will change his life. He suspects that within several hundred years, if this sort of thing continues, the AI systems will control everything. Bob probably doesn’t hesitate, in the way that businesses don’t hesitate to use gas vehicles even if the people involved genuinely think that climate change will be a massive catastrophe in hundreds of years.

When the business strategizing AI systems finally plough all of the resources in the universe into a host of thriving 21st Century businesses, was this misuse or misalignment or accident? The strange new values that were satisfied were those of the AI systems, but the entire outcome only happened because people like Bob chose it knowingly (let’s say). Bob liked it more than the long glorious human future where his business was less good. That sounds like misuse. Yet also in a system of many people, letting this decision fall to Bob may well have been an accident on the part of others, such as the technology’s makers or legislators.

Outcomes are the result of the interplay of choices, driven by different values. Thus it isn’t necessarily sensical to think of them as flowing from one entity’s values or another’s. Here, AI technology created a better option for both Bob and some newly-minted misaligned AI values that it also created—‘Bob has a great business, AI gets the future’—and that option was worse for the rest of the world. They chose it together, and the choice needed both Bob to be a misuser and the AI to be misaligned. But this isn’t a weird corner case, this is a natural way for the future to be destroyed in an economy.

Thanks to Joe Carlsmith for conversation leading to this post.

Energy efficiency of The Spirit of Butt’s Farm

Wed, 18 Nov 2020 23:53:25 GMT

Updated Nov 18, 2020

The Spirit of Butt’s Farm:

covered around 31.67 m/kJ
and moved mass at around 0.16 – 0.086 kg.m/J

Details

The Spirit of Butt’s Farm was a record setting model airplane that crossed the Atlantic on one gallon of fuel.¹ Fully fueled it weighed 4.987 kg, dry it weighed 2.705 kg.²

The record setting flight used 117.1 fluid ounces of fuel.³ The straight line distance of the flight was 3,028.1 km.⁴ It was powered by 88% Coleman lantern fuel, mixed with lubricant.⁵ Coleman fuel is based on naphtha ⁶, so we can use the energy density of naphtha—31.4 MJ/L⁷—as a rough guide to its energy content, though naphtha appears to vary in its content, and it is unclear whether Coleman fuel consists entirely of naphtha.

From all this, we have:

Distance per energy = 3,028.1 km / (117.1 fl oz * 0.88 * 31.4 MJ/L)

= 31.67 m/kJ

For weight times distance per energy we will calculate a best and a worst score. To calculate the best score we will use the fully fueled weight, and to calculate the worst score we will use the dry weight. All other values are the same in both calculations.

Best score:

Distance*mass/energy = 4.987 kg * 31.67 m/kJ

= 0.16 kg.m/J

(4.987 kg * 3,028.1km) / (117.1 US fluid ounces * 31.4MJ/litre) = 0.1389 kg*m/j

Worst score:

Distance*mass/energy = 2.705 kg * 31.67 m/kJ

= .086 kg.m/J

Notes

Photo by Ronan Coyne, licensed under the Creative Commons Attribution-Share Alike 3.0 Unported license, unaltered.

Was the industrial revolution a drastic departure from historic trends?

Tue, 17 Nov 2020 22:14:30 GMT

We do not have a considered view on this topic.

Details

We have not investigated this topic. This is an incomplete list of evidence that we know of:

David Roodman’s analysis of the surprisingness of the industrial revolution under his 2020 model of economic history.¹
Ben Garfinkel’s analysis of whether economic history suggests a singularity²
Robin Hanson’s analysis of historic economic growth understood as a sequence of exponential modes.³
On a log(GWP)-log(doubling time) graph, the industrial revolution appears to be almost perfectly on trend, according to our very crude analysis.

Relevance

The nature of the industrial revolution is relevant to AI forecasting in the following ways:

If growth during the industrial revolution is a highly improbable aberration from longer term trends, it suggests that it is a consequence of specific developments at the time, most saliently new technologies. This suggests that new technologies can sometimes alone cause changes at the level of the global economy.
‘The impact of the industrial revolution’ is sometimes used as a measure against which to compare consequences of AI developments. Thinking here may be sharpened by clarification on the nature of the industrial revolution. This use is likely related to the point above, where ‘the scale of the industrial revolution’ is taken to be a historically plausible scale of impact for the most ambitious technologies.
If economic history is best understood as a sequence of ‘growth modes’, per Hanson 2000,⁴ the industrial revolution being one, this changes our best extrapolation to the future. For instance, we might expect the continuation of the current mode to be slower than in a continuously super-exponential model, but may also expect to meet a new growth mode at some point, which may be substantially faster (and have other characteristics recognizable from past ‘growth mode’ changes). See Hanson 2000 for more on this.

Notes

Energy efficiency of MacCready Gossamer Albatross

Tue, 10 Nov 2020 02:15:26 GMT

Updated Nov 9, 2020

The MacCready Gossamer Albatross:

covered around 2.0—4.6 m/kJ
and moved mass at around 0.1882 —0.4577 kg.m/J

Details

The MacCready Gossamer Albatross was a human-powered flying machine that crossed the English Channel in 1979.¹ The pilot pedaled the craft, seemingly as if on a bicycle. It had a gross mass of 100kg, flying across the channel,² and flew 35.7 km in 2 hours and 49 minutes.³ The crossing was difficult however, so it seems plausible that the Gossamer Albatross could fly more efficiently in better conditions.

We do not know the pilot’s average power output, however:

Wikipedia claims at least 300W was required to fly the craft⁴
Chung 2006, an engineering textbook, claims that the driver, a cyclist, could produce around 200W of power.⁵
Our impression is that 200W is a common power output over houres for amateur cycling. For instance, one of our researchers is able to achieve this for three hours.⁶

The best documented human cycling wattage that we could easily find is from professional rider Giulio Ciccone who won a stage of the Tour de France, then uploaded power data to the fitness tracking site Strava.⁷ His performance suggests around 318W is a reasonable upper bound, supposing that the pilot of the Gossamer Albatross would have had lower performance.⁸

To find the energy used by the cyclist, we divided power output by typical efficiency for a human on a bicycle, which according to Wikipedia ranges from .18 to .26.⁹

Distance per Joule

For distance per energy this gives us a highest measure of:

35.7 km / ((200W * (2 hours + 49 minutes))/0.26) = 4,577 m/MJ

And a lowest measure of:

35.7 km / ((318W * (2 hours + 49 minutes))/0.18) = 1,993 m/MJ

Mass per Joule

For weight times distance per energy this gives us a highest measure of:

(100kg * 35.7 km) / ((200W * (2 hours + 49 minutes))/0.26) = 0.4577 kg*m/j

And a lowest measure of:

(100kg * 35.7 km) / ((318W * (2 hours + 49 minutes))/0.17) = 0.1882 kg*m/j

Notes

Energy efficiency of Boeing 747-400

Fri, 06 Nov 2020 05:10:05 GMT

Updated Nov 5, 2020

The Boeing 747-400:

covers around 0.0021m/kJ.
and moves mass at around 0.39 – 0.83 kg.m/J

Details

The Boeing 747-400 is a 1987 passenger plane.¹

Distance per Joule

The plane uses 10.77 kg/km of fuel on a medium haul flight.² We do not know what type of fuel it uses, but typical values for aviation fuel are around 44MJ/kg.³ Thus to fly a kilometer, the plane needs 10.77 kg of fuel, which is 10.77 x 44 MJ = 474 MJ of fuel. This gives us 0.0021m/kJ.

Mass.distance per Joule

According to Wikipedia, the 747’s ‘operating empty weight’ is 183,523 kg and its ‘maximum take-off weight’ is 396,893 kg.⁴ We use the range 183,523 kg—396,893 kg since we do not know at what weight in that range the relevant speeds were measured.

We have:

Distance per kilojoule: 0.0021m/kJ
Mass: 183,523 kg—396,893 kg

This gives us a range of 0.39 – 0.83 kg.m/J

Notes

Energy efficiency of Airbus A320

Fri, 06 Nov 2020 04:44:03 GMT

Updated Nov 5, 2020

The Airbus A320:

covers around 0.0078 m/kJ
and moves mass at around 0.33 – 0.61 kg.m/J

Details

The Airbus A320 is a 1987 passenger plane.¹

Distance per Joule

The plane uses 2.91 kg of fuel per km on a medium haul flight.² We do not know what type of fuel it uses, but typical values for aviation fuel are around 44MJ/kg.³ Thus to fly a kilometer, the plane needs 2.91kg of fuel, which is 2.91 x 44 MJ = 128MJ of fuel. This gives us 0.0078 m/kJ

Mass.distance per Joule

According to modernairliners.com, the A320’s ‘operating empty weight’ is 42,600 kg and its ‘maximum take-off weight’ is 78,000 kg.⁴ We use the range 42,600—78,000 kg, since we do not know at what weight in that range the relevant speeds were measured.

We have:

Distance per kilojoule: 0.0078 m/kJ
Mass: 42,600—78,000 kg

This gives us a range of 0.33 – 0.61 kg.m/J

Notes

Energy efficiency of North American P-51 Mustang

Fri, 06 Nov 2020 01:04:24 GMT

Updated Nov 5, 2020

The North American P-51 Mustang:

flew around 0.073—0.092 m/kJ
and moved mass at around 0.25 – 0.50 kg.m/J

Details

The North American P-51 Mustang was a 1940 US WWII fighter and fighter-bomber.¹

Mass

According to Wikipedia²:

Empty weight: 7,635 lb (3,465 kg)
Gross weight: 9,200 lb (4,175 kg)
Max takeoff weight: 12,100 lb (5,488 kg)

We use the range 3,465—5,488 kg, since we do not know at what weight in that range the relevant speeds were measured.

Distance per Joule

Wikipedia tells us that cruising speed was 362 mph (162 m/s)³

A table from WWII Aircraft Performance gives combinations of flight parameters apparently for a version of the P-51, however it has no title or description, so we cannot be confident. ⁴ We extracted some data from it here. This data suggests the best combination of parameters gives a fuel economy of 6.7 miles/gallon (10.8km)

We don’t know what fuel was used, but fuel energy density seems likely to be between 31—39 MJ/L = 117—148 MJ/gallon.⁵

Thus the plane flew about 10.8km on 117—148 MJ of fuel, for 0.073—0.092 m/kJ

Mass.distance per Joule

We have:

Distance per kilojoule: 0.073—0.092 m/kJ
Mass: 3,465—5,488 kg

This gives us a range of 0.25 – 0.50 kg.m/J

Notes

Energy efficiency of Vickers Vimy plane

Thu, 05 Nov 2020 20:54:31 GMT

Updated Nov 5, 2020

The Vickers Vimy:

flew around 0.025—0.050 m/kJ
and moved mass at around 0.081 – 0.25 kg.m/J

Details

The Vickers Vimy was a 1917 British WWI bomber.¹ It was used in the first non-stop transatlantic flight.

Mass

According to Wikipedia²:

Empty weight: 7,104 lb (3,222 kg)
Max takeoff weight: 10,884 lb (4,937 kg)

We use the range 3,222—4,937 kg, since we do not know at what weight in that range the relevant speeds were measured.

Energy use per second

We also have:

Power: 360 horsepower = 270 kW³
Efficiency of use of energy from fuel: we did not find data on this, so use an estimate of 15%-30%, based on what we know about the energy efficiency of the Wright Flyer.

From these we can calculate:

Energy use per second
= power of engine x 1/efficiency in converting energy to engine power
= 270kJ/s / .15—270kJ/s / .30
= 900—1800 kJ/s

Distance per second

Wikipedia gives us:

Maximum speed: 100 mph (160 km/h, 87 kn)⁴

Note that the figures for power do not obviously correspond to the highest measured speed. This is a rough estimate.

Distance per Joule

We now have (from above):

speed = 100 miles/h = 44.7m/s
energy use = 900—1800 kJ/s

Thus, on average each second the plane flies 44.7 m and uses 900—1800 kJ, for 0.025—0.050 m/kJ.

Mass.distance per Joule

We have:

Distance per kilojoule: 0.025—0.050 m/kJ
Mass: 3,222—4,937 kg

This gives us a range of 0.081 – 0.25 kg.m/J

Notes

Energy efficiency of Wright model B

Thu, 05 Nov 2020 20:09:52 GMT

Updated Nov 5, 2020

The Wright model B:

flew around 0.10—0.21m/kJ
and moved mass at around 0.036 – 0.12 kg.m/J

Details

The Wright Model B was a 1910 plane developed by the Wright Brothers.¹

Mass

According to Wikipedia²:

Empty weight: 800 lb (363 kg)
Gross weight: 1,250 lb (567 kg)

We use the range 363—567 kg, since we do not know at what weight in that range the relevant speeds were measured.

Energy use per second

From Wikipedia, we have³:

Power: 35 horsepower = 26kW
Efficiency of use of energy from fuel: we could not find data on this, so use an estimate of 15%-30%, based on what we know about the energy efficiency of the Wright Flyer.

From these we can calculate:

Energy use per second
= power of engine x 1/efficiency in converting energy to engine power
= 26kJ/s / .15—26kJ/s / .30
= 86.6—173 kJ/s

Distance per second

Wikipedia gives us⁴:

Maximum speed: 45 mph (72 km/h, 39 kn)
Cruise speed: 40 mph (64 km/h, 35 kn)

We use the cruise speed, as it seems more likely to represent speed achieved with the energy usages reported.

Distance per Joule

We now have (from above):

speed = 40miles/h = 17.9m/s
energy use = 86.6—173 kJ/s

Thus, on average each second the plane flies 17.9m and uses 86.6—173 kJ, for 0.10—0.21m/kJ.

Mass.distance per Joule

We have:

Distance per kilojoule: 0.10—0.21m/kJ
Mass: 363—567kg

This gives us a range of 0.036 – 0.12 kg.m/J

Notes

Energy efficiency of Wright Flyer

Wed, 04 Nov 2020 18:58:55 GMT

The Wright Flyer:

flew around 0.08-0.17m/kJ
and moved mass at around .02 – .06 kg.m/J

Details

The Wright Flyer (Flyer I) was the first successful plane, built in 1903.¹

Mass

According to Wikipedia²:

Empty weight: 605 lb (274 kg)
Max takeoff weight: 745 lb (338 kg)

Energy use per second

Fuel use per hour

A 1904 article in the Minneapolis Journal says the plane consumed ‘a little less than than ten pounds of gasoline per hour’.³ A pound of gasoline contains around 20MJ⁴ So we have:

Hourly fuel consumption: 10lb/h x 20MJ/lb = 200MJ/h = 55kJ/s

We don’t know how reliable this source is. For instance, a 1971 book, The Write Brothers’ Engines and their Design does not give data on fuel consumption in their table of engine characteristics for lack of available comprehensive data⁵, suggesting that its authors did not consider the article strong evidence, though it is also possible that they didn’t have access to the 1904 article.

Utilized motor power / efficiency

To confirm, we can estimate the plane’s energy use per second by a second means: combining the claimed power (energy/second) made use of by the engine, and a guess about how much fuel is needed to deliver that amount of energy.

Utilized motor power

According to Wikipedia, the plane had a 12 horsepower (9 kJ/s), gasoline engine.⁶ The 1904 Minneapolis Journal article put it at 16 horsepower (12 kJ/s), and Orville Wright, quoted by Hobbs (1971), puts it at ‘almost 16 horsepower’ at one point.⁷ Hobbs says that at one point this engine achieved 25 horsepower, though this probably isn’t representative of what was ‘actually utilized’.⁸ For that he gives a range of 8.25-16 horsepower.⁹ In light of these estimates, we will use 8.25-16 horsepower, which is 6.15-12 kJ/s.

Efficiency of energy conversion from fuel to motor power

A quote from Orville Wright suggests fuel consumption as 0.580lb of fuel per horsepower hour.¹⁰ This would imply 23% of energy was utilized from the fuel.¹¹

Hobbs notes that this seems low, but assumes a similar efficiency: 24.50%. Thus he presumably doesn’t find it implausible.¹²

According to Wikipedia, the thermal efficiency of a typical gasoline engine is 20%¹³. It seems that this has increased¹⁴, which would suggest that the typical figure was lower in 1903. We don’t think this undermines the more specific figures given above.

It seems likely that Hobbs number is best here, since he knows about Wright’s number, and may have more information than us about for instance the exact fuel being used. So we use 24.5%.

Calculation of energy use from motor power/efficiency

Combining these numbers, we have:

power spent = 6.15-12 kJ/s used by engine x 1 / 24.5% fuel energy needed to get one unit of energy used by engine, given inefficiency

= 25-49 kJ/s

Calculation of energy use

We now have:

energy expenditure calculated via motor power and efficiency: 25-49 kJ/s
energy expenditure calculated via hourly fuel use: 55kJ/s

Neither figure seems clearly more reliable, so we will use the range 25-55kJ/s

Distance per second

Two of the Wright Flyer’s first flights were 120 feet in 12 seconds and 852 feet in 59 seconds.¹⁵ This gives speeds of 3m/s and 4.4m/s. We will use the second, since it better represents successful flight, still within the first days.

Distance per Joule

We now have (from above):

speed = 4.4 m/s
energy use = 25-55 kJ/s

Thus, on average each second the plane flies 4.4m and uses 25-55kJ, for 0.08-0.17m/kJ.

Mass.distance per Joule

We have:

Distance per kilojoule: 0.08-0.17m/kJ
Mass: 274-338kg

This gives us a range of .02 – .06 kg.m/J

Automated intelligence is not AI

Sun, 01 Nov 2020 23:38:44 GMT

By Katja Grace, Nov 1 2020, Crossposted from world spirit sock puppet.

Sometimes we think of ‘artificial intelligence’ as whatever technology ultimately automates human cognitive labor.

I question this equivalence, looking at past automation. In practice human cognitive labor is replaced by things that don’t seem at all cognitive, or like what we otherwise mean by AI.

Some examples:

Early in the existence of bread, it might have been toasted by someone holding it close to a fire and repeatedly observing it and recognizing its level of doneness and adjusting. Now we have machines that hold the bread exactly the right distance away from a predictable heat source for a perfect amount of time. You could say that the shape of the object embodies a lot of intelligence, or that intelligence went into creating this ideal but non-intelligent tool.
Self-cleaning ovens replace humans cleaning ovens. Humans clean ovens with a lot of thought—looking at and identifying different materials and forming and following plans to remove some of them. Ovens clean themselves by getting very hot.
Carving a rabbit out of chocolate takes knowledge of a rabbit’s details, along with knowledge of how to move your hands to translate such details into chocolate with a knife. A rabbit mold automates this work, and while this route may still involve intelligence in the melting and pouring of the chocolate, all rabbit knowledge is now implicit in the shape of the tool, though I think nobody would call a rabbit-shaped tin ‘artificial intelligence’.
Human pouring of orange juice into glasses involves various mental skills. For instance, classifying orange juice and glasses and judging how they relate to one another in space, and moving them while keeping an eye on this. Automatic orange juice pouring involves for instance a button that can only be pressed with a glass when the glass is in a narrow range of locations, which opens an orange juice faucet running into a spot common to all the possible glass-locations.

Some of this is that humans use intelligence where they can use some other resource, because it is cheap on the margin where the other resource is expensive. For instance, to get toast, you could just leave a lot of bread at different distances then eat the one that is good. That is bread-expensive and human-intelligence-cheap (once you come up with the plan at least). But humans had lots of intelligence and not much bread. And if later we automate a task like this, before we have computers that can act very similarly to brains, then the alternate procedure will tend to be one that replaces human thought with something that actually is cheap at the time, such as metal.

I think a lot of this is that to deal with a given problem you can either use flexible intelligence in the moment, or you can have an inflexible system that happens to be just what you need. Often you will start out using the flexible intelligence, because being flexible it is useful for lots of things, so you have some sitting around for everything, whereas you don’t have an inflexible system that happens to be just what you need. But if a problem seems to be happening a lot, it can become worth investing the up-front cost of getting the ideal tool, to free up your flexible intelligence again.

Time for AI to cross the human range in English draughts

Mon, 26 Oct 2020 22:28:36 GMT

Updated 26 Oct 2020

AI progress in English draughts performance crossed the following ranges in the following times:

Range	Start	End	Duration (years)
First attempt to beginner level	1951	~1956, <1961	~4, <10
Beginner to superhuman	~1956, <1961	1994	~38, >33
Above superhuman	1994	2007*	13*

* treating perfect play as the end of progress, though progress could potentially be made in performing better against imperfect play.

Details

Metric

‘English Draughts’ is a popular variety of draughts, or checkers.

Here we look at direct success of AI in beating human players, rather than measuring humans and AI on a separate metric of strength.

Data

Data here comes mostly from Wikipedia. We found several discrepancies in Wikipedia’s accounts of these events, so consider the remaining data to be somewhat unreliable.

AI achievement of human milestones

Earliest attempt

According to Wikipedia, the first checkers program was run in 1951.¹

Beginner level

There seems to be some ambiguity around the timing and performance of Arthur Samuel’s early draughts programs, but it appears that he worked on them from around 1952. In 1956, he demonstrated a program on television. ² It is unclear how good the program’s play was, but it is said to have resulted in a fifteen-point rise in the stock price of IBM when demonstrated to IBM shareholders, seemingly at around that time. This weakly suggests that the program played at at least beginner level.

In 1962 Samuel’s program apparently beat an ambiguously skilled player who would by four-years later become a state champion in Connecticut.³ Thus progress was definitely beyond beginner level by 1962.

Superhuman level

In 1994, computer program Chinook drew six times against world champion Marius Tinsley, before Tinsley withdrew due to pancreatic cancer and Chinook officially won.⁴ Thus Chinook appears to have been close to as good as the best player in 1994.

End of progress

In 2007 checkers was ‘weakly solved’, which is to say that perfect play guaranteeing a draw for both sides from the start of the game is known, from the starting state⁵ (this does not imply that if someone plays imperfectly, perfect moves following this are known).⁶ This is not the best possible performance by all measures, since further progress could presumably be made on reliably beating worse players.

Times for AI to cross human-relative ranges

Given the above dates, we have:

Range	Start	End	Duration (years)
First attempt to beginner level	1951	~1956, <1961	~4, <10
Beginner to superhuman	~1956, <1961	1994	~38, >33
Above superhuman	1994	2007*	13*

* treating perfect play as the end of progress, though progress could potentially be made in performing better against imperfect play.

Notes

Time for AI to cross the human range in StarCraft

Tue, 20 Oct 2020 20:55:24 GMT

Updated 22 Oct 2020

Progress in AI StarCraft performance took:

~0 years to reach the level of an untrained human
~21 years to pass from untrained human level to high professional human level
~2 years to continue from trained human to current performance (2020), with no particular end in sight.

Details

Metric

We compare human and AI players on their direct ability to beat one another (rather than a measure of the overall performance of each).

AI milestones

Earliest attempt

Starcraft was released in 1998¹ The game allows the player to play against a computer opponent, however this built-in AI has access to information that a normal player would not have. For instance, it has real-time information about what the other player is doing at all times, which is normally hidden. We do not have detailed knowledge about early StarCraft AIs that do not have this advantage, but our impression is that it was possible to write them from the start (see next section).

Beginner level

Our impression is that since StarCraft Brood War came out in 1998, it has been possible to write a bot that can beat a player who recently learned the game, without “cheating” in the way that the game’s built-in computer opponents do. This is uncertain, and based on private discussion with people who write Starcraft AIs that compete in tournaments.

Professional level

In 2018, DeepMind’s AlphaStar beat MaNa², a strong professional player (seemingly 13th place in the 2018 StarCraft II World Championship Series Circuit)³. This does not imply that AlphaStar was in general a stronger player than MaNa, but suggests AlphaStar was at a broadly high professional level. How AlphaStar’s performance compares to humans will depend on how narrowly the task is defined, so that if the AI is not allowed to give commands faster than a human is able, it will compare less favorably than if it is allowed to give commands very quickly⁴.

Times for AI to cross human-relative ranges

Given the above dates, we have:

Range	Start	End	Duration (years)
First attempt to beginner level	1998	1998	~0
Beginner to superhuman	1998	2018	~21
Above superhuman	2018	?	>2

Notes

Time for AI to cross the human performance range in ImageNet image classification

Mon, 19 Oct 2020 23:52:50 GMT

Time for AI to cross the human performance range in Go

Fri, 16 Oct 2020 00:05:43 GMT

It took over 30 years for AI to progress from human beginner level to superhuman level.

Details

Human performance milestones

Human go ratings range from 30 kyu (beginner), through 7 dan to at least 9 professional dan.¹ These ratings go downwards through kyu levels, then upward through dan levels, then upward through professional dan levels. The top ratings seem to be closer together than the lower ones, though there are apparently multiple systems which vary)²

AI achievement of human milestones

Earliest attempt

Wikipedia says the first Go program was written in 1968.³ We do not know how well it performed.

Beginner level

We have not investigated early Go performance in depth. Figure 1 includes informed guesses about early performance by David Fotland, author of successful Go program, The Many Faces of Go, and Sensei’s Library, a Go wiki.⁴ Fotland says that early data on AI Go performance is poor, since bots did not play in tournaments, so were not rated.

Figure 1: From Grace 2013.

This suggests that by 1987 Go bots were performing better than human beginners. We do not have evidence to pin down the date of human beginner level AI better, but have also not investigated thoroughly (there appears to be more evidence).

Superhuman level

In May 2017 AlphaGo beat the top ranked Go player in the world.⁵ This does not imply that AlphaGo was overall better, but a new version in October could beat the May version in 89 games out of 100⁶, suggesting that if in May it would have beaten Ke Jie in more than 11% of games, the new version would beat Ke Jie more than half the time, i.e. perform better than the best human player. Thus 2017 seems like a reasonable date for top human-level play.

Times for AI to cross human-relative ranges

Given the above dates, we have:

Range	Start	End	Duration (years)
First attempt to beginner level	1968	<1987	<19
Beginner to superhuman	<1987	2017	>30
Above superhuman	2017	>2020	>3

Time for AI to cross the human performance range in chess

Tue, 13 Oct 2020 23:36:30 GMT

Updated 15 Oct 2020

Progress in computer chess performance took:

~0 years to go from from playing chess at all to playing it at human beginner level
~49 years to go from human beginner level to superhuman level
~11 years to go from superhuman level to the the current highest performance

Details

Human range performance milestones

We use the common Elo system for measuring chess performance. Human chess Elo ratings range from around 800 (beginner)¹ to 2882 (highest recorded).² The highest recorded human score is likely higher than it would have been without chess AI existing, since top players can learn from the AI.³

Times for machines to cross ranges

Beginner to superhuman range

We could not find transparent sources for low computer chess Elo records, but it seems common to place Elo scores of 800-1200 in the 1950s and 1960s. In his book Robot (1999)[/note]Moravec, Hans. Robot: Mere Machine to Transcendent Mind, n.d., p71, also at https://frc.ri.cmu.edu/~hpm/book97/ch3/index.html[/note], Moravec gives the diagram shown in Figure 1, which puts a machine with an Elo of around 800 in 1957. He does not appear to provide a source for this however. Figure 2 shows another figure without sources from a 2002 article by L. Stephen Coles at Dr. Dobbs ⁴, which puts some machine at over 1000 in around 1950. To err on the side of assuming narrow human ranges, and because Moravec appears to be a more reliable source, we use his data here. This means that machine chess performance entered the human range in 1957 at the latest.

Figure 1: Graph from Moravec, 1999⁵. We did not find Moravec’s sources for these numbers.

Figure 2: Chess AI progress compared to human performance, from Coles 2002⁶. The image appears to confusingly claim that the present time is around 1993, in which case the right of the graph (after ‘now’) must be imagined, though it appears to be approximately correct.

The chess computer Deep Blue famously beat the then world-champion Kasparov under tournament conditions in 1997.⁷ However this does not imply that Deep Blue was at that point overall more capable than Kasparov, i.e. had a higher Elo rating.⁸

According to the Swedish Chess Computer Association records, 2006 is the year when the highest machine Elo rating surpassed the highest human Elo (both the highest at the time, and the highest in 2020). In particular Rybka 1.2 was rated 2902. At the time, the highest human Elo rating was Garry Kasparov at 2851.⁹

Thus it took around 49 years for computers to progress from beginner human level chess to superhuman chess.

Pre-human range

The Chess Programming Wiki says that the 1957 Bernstein Chess program was the first complete chess program.¹⁰ This seems likely to be the same Bernstein program noted by Moravec as having an 800 Elo in 1957 (see above). Thus if correct, this means that once machines could complete the task of playing chess at all, they could already do it at human beginner level. This may not be accurate (none of these sources appear to be very reliable), but it strongly suggests that the time between lowest possible performance and beginner human performance was not as long as decades.

Superhuman performance range

The Swedish Chess Computer Association has measured continued progress. As of July 2020, the best chess machine is rated 3558¹¹, whereas in 2019 sometime, the highest rating was 3529.¹² Alphazero also appeared to have an Elo just below 3500 in 2017, according to its creators (from a small figure with unclear labels).¹³

We know of no particular upper bound to chess performance.

This suggests that so far the superhuman range in chess playing has permitted at least least 14 years of further progress, and may permit much more.

Notes

Relevant pre-AGI possibilities

Fri, 19 Jun 2020 13:40:00 GMT

By Daniel Kokotajlo¹

Epistemic status: I started this as an AI Impacts research project, but given that it’s fundamentally a fun speculative brainstorm, it worked better as a blog post.

The default, when reasoning about advanced artificial general intelligence (AGI), is to imagine it appearing in a world that is basically like the present. Yet almost everyone agrees the world will likely be importantly different by the time advanced AGI arrives.

One way to address this problem is to reason in abstract, general ways that are hopefully robust to whatever unforeseen developments lie ahead. Another is to brainstorm particular changes that might happen, and check our reasoning against the resulting list.

This is an attempt to begin the second approach.² I sought things that might happen that seemed both (a) within the realm of plausibility, and (b) probably strategically relevant to AI safety or AI policy.

I collected potential list entries via brainstorming, asking others for ideas, googling, and reading lists that seemed relevant (e.g. Wikipedia’s list of emerging technologies,³ a list of Ray Kurzweil’s predictions⁴, and DARPA’s list of projects.⁵)

I then shortened the list based on my guesses about the plausibility and relevance of these possibilities. I did not put much time into evaluating any particular possibility, so my guesses should not be treated as anything more. I erred on the side of inclusion, so the entries in this list vary greatly in plausibility and relevance. I made some attempt to categorize these entries and merge similar ones, but this document is fundamentally a brainstorm, not a taxonomy, so keep your expectations low.

I hope to update this post as new ideas find me and old ideas are refined or refuted. I welcome suggestions and criticisms; email me (gmail kokotajlod) or leave a comment.

Interactive “Generate Future” button

Asya Bergal and I made an interactive button to go with the list. The button randomly generates a possible future according to probabilities that you choose. It is very crude, but it has been fun to play with, and perhaps even slightly useful. For example, once I decided that my credences were probably systematically too high because the futures generated with them were too crazy. Another time I used the alternate method (described below) to recursively generate a detailed future trajectory, written up here. I hope to make more trajectories like this in the future, since I think this method is less biased than the usual method for imagining detailed futures.⁶

To choose probabilities, scroll down to the list below and fill each box with a number representing how likely you think the entry is to occur in a strategically relevant way prior to the advent of advanced AI. (1 means certainly, 0 means certainly not. The boxes are all 0 by default.) Once you are done, scroll back up and click the button.

A major limitation is that the button doesn’t take correlations between possibilities into account. The user needs to do this themselves, e.g. by redoing any generated future that seems silly, or by flipping a coin to choose between two generated possibilities that seem contradictory, or by choosing between them based on what else was generated.

Here is an alternate way to use this button that mostly avoids this limitation:

Fill all the boxes with probability-of-happening-in-the-next-5-years (instead of happening before advanced AGI, as in the default method)
Click the “Generate Future” button and record the results, interpreted as what happens in the next 5 years.
Update the probabilities accordingly to represent the upcoming 5-year period, in light of what has happened so far.
Repeat steps 2 – 4 until satisfied. I used a random number generator to determine whether AGI arrived each year.

If you don’t want to choose probabilities yourself, click “fill with pre-set values” to populate the fields with my non-expert, hasty guesses.⁷

Key

Letters after list titles indicate that I think the change might be relevant to:

TML: Timelines—how long it takes for advanced AI to be developed
TAS: Technical AI safety—how easy it is (on a technical level) to make advanced AI safe, or what sort of technical research needs to be done
POL: Policy—how easy it is to coordinate relevant actors to mitigate risks from AI, and what policies are relevant to this.
CHA: Chaos—how chaotic the world is.⁸
MIS: Miscellaneous

Each possibility is followed by some explanation or justification where necessary, and a non-exhaustive list of ways the possibility may be relevant to AI outcomes in particular (which is not guaranteed to cover the most important ones). Possibilities are organized into loose categories created after the list was generated.

List of strategically relevant possibilities

Inputs to AI

1. Advanced science automation and research tools (TML, TAS, CHA, MIS)

Narrow research and development tools might speed up technological progress in general or in specific domains. For example, several of the other technologies on this list might be achieved with the help of narrow research and development tools.

2. Dramatically improved computing hardware (TML, TAS, POL, MIS)

By this I mean computing hardware improves at least as fast as Moore’s Law. Computing hardware has historically become steadily cheaper, though it is unclear whether this trend will continue. Some example pathways by which hardware might improve at least moderately include:

Ordinary scale economies⁹
Improved data locality¹⁰
Increased specialization for specific AI applications¹¹
Optical computing¹²
Neuromorphic chips¹³
3D integrated circuits¹⁴
Wafer-scale chips¹⁵
Quantum computing¹⁶
Carbon nanotube field-effect transistors¹⁷

Dramatically improved computing hardware may:

Cause any given AI capability to arrive earlier
Increase the probability of hardware overhang.
Affect which kinds of AI are developed first (e.g. those which are more compute-intensive.)
Affect AI policy, e.g. by changing the relative importance of hardware vs. research talent

3. Stagnation in computing hardware progress (TML, TAS, POL, MIS)

Many forecasters think Moore’s Law will be ending soon (as of 2020).¹⁸ In the absence of successful new technologies, computing hardware could progress substantially more slowly than Moore’s Law would predict.

Stagnation in computing hardware progress may:

Cause any given AI capability to arrive later
Decrease the probability of hardware overhang.
Affect which kinds of AI are developed first (e.g. those which are less compute-intensive.)
Influence the relative strategic importance of hardware compared to researchers
Make energy and raw materials a greater part of the cost of computing

4. Manufacturing consolidation (POL)

Chip fabrication has become more specialized and consolidated over time, to the point where all of the hardware relevant to AI research depends on production from a handful of locations.¹⁹ Perhaps this trend will continue.

One country (or a small number working together) could control or restrict AI research by controlling the production and distribution of necessary hardware.

5. Advanced additive manufacturing (e.g. 3D printing or nanotechnology) (TML, CHA)

Advanced additive manufacturing could lead to various materials, products and forms of capital being cheaper and more broadly accessible, as well as to new varieties of them becoming feasible and quicker to develop. For example, sufficiently advanced 3D printing could destabilize the world by allowing almost anyone to secretly produce terror weapons. If nanotechnology advances rapidly, so that nanofactories can be created, the consequences could be dramatic:²⁰

Greatly reduced cost of most manufactured products
Greatly faster growth of capital formation
Lower energy costs
New kinds of materials, such as stronger, lighter spaceship hulls
Medical nanorobots
New kinds of weaponry and other disruptive technologies

6. Massive resource glut (TML, TAS, POL, CHA)

By “glut” I don’t necessarily mean that there is too much of a resource. Rather, I mean that the real price falls dramatically. Rapid decreases in the price of important resources have happened before.²¹ It could happen again via:

Cheap energy (e.g. fusion power, He-3 extracted from lunar regolith,²² methane hydrate extracted from the seafloor,²³ cheap solar energy²⁴)
A source of abundant cheap raw materials (e.g. asteroid mining,²⁵ undersea mining²⁶)
Automation of relevant human labor. Where human labor is an important part of the cost of manufacturing, resource extraction, or energy production, automating labor might substantially increase economic growth, which might result in a greater amount of resources devoted to strategically relevant things (such as AI research) which is relevantly similar to a price drop even if technically the price doesn’t drop.²⁷ and therefore investment in AI.

My impression is that energy, raw materials, and unskilled labor combined are less than half the cost of computing, so a decrease in the price of one of these (and possibly even all three) would probably not have large direct consequences on the price of computing.²⁸ But a resource glut might lead to general economic prosperity, with many subsequent effects on society, and moreover the cost structure of computing may change in the future, creating a situation where a resource glut could dramatically lower the cost of computing.²⁹

7. Hardware overhang (TML, TAS, POL)

Hardware overhang refers to a situation where large quantities of computing hardware can be diverted to running powerful AI systems as soon as the AI software is developed.

If advanced AGI (or some other powerful software) appears during a period of hardware overhang, its capabilities and prominence in the world could grow very quickly.

8. Hardware underhang (TML, TAS, POL)

The opposite of hardware overhang might happen. Researchers may understand how to build advanced AGI at a time when the requisite hardware is not yet available. For example, perhaps the relevant AI research will involve expensive chips custom-built for the particular AI architecture being trained.

A successful AI project during a period of hardware underhang would not be able to instantly copy the AI to many other devices, nor would they be able to iterate quickly and make an architecturally improved version.

Technical tools

9. Prediction tools (TML, TAS, POL, CHA, MIS)

Tools may be developed that are dramatically better at predicting some important aspect of the world; for example, technological progress, cultural shifts, or the outcomes of elections, military clashes, or research projects. Such tools could for instance be based on advances in AI or other algorithms, prediction markets, or improved scientific understanding of forecasting (e.g. lessons from the Good Judgment Project).

Such tools might conceivably increase stability via promoting accurate beliefs, reducing surprises, errors or unnecessary conflicts. However they could also conceivably promote instability via conflict encouraged by a powerful new tool being available to a subset of actors. Such tools might also help with forecasting the arrival and effects of advanced AGI, thereby helping guide policy and AI safety work. They might also accelerate timelines, for instance by assisting project management in general and notifying potential investors when advanced AGI is within reach.

10. Persuasion tools (POL, CHA, MIS)

Present technology for influencing a person’s beliefs and behavior is crude and weak, relative to what one can imagine. Tools may be developed that more reliably steer a person’s opinion and are not so vulnerable to the victim’s reasoning and possession of evidence. These could involve:

Advanced understanding of how humans respond to stimuli depending on context, based on massive amounts of data
Coaching for the user on how to convince the target of something
Software that interacts directly with other people, e.g. via text or email

Strong persuasion tools could:

Allow a group in conflict who has them to quickly attract spies and then infiltrate an enemy group
Allow governments to control their populations
Allow corporations to control their employees
Lead to a breakdown of collective epistemology³⁰

11. Theorem provers (TAS)

Powerful theorem provers might help with the kinds of AI alignment research that involve proofs or help solve computational choice problems.

12. Narrow AI for natural language processing (TML, TAS, CHA)

Researchers may develop narrow AI that understands human language well, including concepts such as “moral” and “honest.”

Natural language processing tools could help with many kinds of technology, including AI and various AI safety projects. They could also help enable AI arbitration systems. If researchers develop software that can autocomplete code—much as it currently autocompletes text messages—it could multiply software engineering productivity.

13. AI interpretability tools (TML, TAS, POL)

Tools for understanding what a given AI system is thinking, what it wants, and what it is planning would be useful for AI safety.³¹

14. Credible commitment mechanisms (POL, CHA)

There are significant restrictions on which contracts governments are willing and able to enforce–for example, they can’t enforce a contract to try hard to achieve a goal, and won’t enforce a contract to commit a crime. Perhaps some technology (e.g. lie detectors, narrow AI, or blockchain) could significantly expand the space of possible credible commitments for some relevant actors: corporations, decentralized autonomous organizations, crowds of ordinary people using assurance contracts, terrorist cells, rogue AGIs, or even individuals.

This might destabilize the world by making threats of various kinds more credible, for various actors. It might stabilize the world in other ways, e.g. by making it easier for some parties to enforce agreements.

15. Better coordination tools (POL, CHA, MIS)

Technology for allowing groups of people to coordinate effectively could improve, potentially avoiding losses from collective choice problems, helping existing large groups (e.g. nations and companies) to make choices in their own interests, and producing new forms of coordinated social behavior (e.g. the 2010’s saw the rise of the Facebook group)). Dominant assurance contracts,³² improved voting systems,³³ AI arbitration systems, lie detectors, and similar things not yet imagined might significantly improve the effectiveness of some groups of people.

If only a few groups use this technology, they might have outsized influence. If most groups do, there could be a general reduction in conflict and increase in good judgment.

Human effectiveness

16. Deterioration of collective epistemology (TML, TAS, POL, CHA, MIS)

Society has mechanisms and processes that allow it to identify new problems, discuss them, and arrive at the truth and/or coordinate a solution. These processes might deteriorate. Some examples of things which might contribute to this:

Increased investment in online propaganda by more powerful actors, perhaps assisted by chatbots, deepfakes and persuasion tools
Echo chambers, filter bubbles, and online polarization, perhaps driven in part by recommendation algorithms
Memetic evolution in general might intensify, increasing the spreadability of ideas/topics at the expense of their truth/importance³⁴
Trends towards political polarization and radicalization might exist and continue
Trends towards general institutional dysfunction might exist and continue

This could cause chaos in the world in general, and lead to many hard-to-predict effects. It would likely make the market for influencing the course of AI development less efficient (see section on “Landscape of…” below) and present epistemic hazards for anyone trying to participate effectively.

17. New and powerful forms of addiction (TML, POL, CHA, MIS)

Technology that wastes time and ruins lives could become more effective. The average person spends 144 minutes per day on social media, and there is a clear upward trend in this metric.³⁵ The average time spent watching TV is even greater.³⁶ Perhaps this time is not wasted but rather serves some important recuperative, educational, or other function. Or perhaps not; perhaps instead the effect of social media on society is like the effect of a new addictive drug — opium, heroin, cocaine, etc. — which causes serious damage until society adapts. Maybe there will be more things like this: extremely addictive video games, or newly invented drugs, or wireheading (directly stimulating the reward circuitry of the brain).³⁷

This could lead to economic and scientific slowdown. It could also concentrate power and influence in fewer people—those who for whatever reason remain relatively unaffected by the various productivity-draining technologies. Depending on how these practices spread, they might affect some communities more or sooner than others.

18. Medicine to boost human mental abilities (TML, CHA, MIS)

To my knowledge, existing “study drugs” such as modafinil don’t seem to have substantially sped up the rate of scientific progress in any field. However, new drugs (or other treatments) might be more effective. Moreover, in some fields, researchers typically do their best work at a certain age. Medicine which extends this period of peak mental ability might have a similar effect.

This could speed up the rate of scientific progress in some fields, among other effects.

19. Genetic engineering, human cloning, iterated embryo selection (TML, POL, CHA, MIS)

Changes in human capabilities or other human traits via genetic interventions³⁸ could affect many areas of life. If the changes were dramatic, they might have a large impact even if only a small fraction of humanity were altered by them.

Changes in human capabilities or other human traits via genetic interventions might:

Accelerate research in general
Differentially accelerate research projects that depend more on “genius” and less on money or experience
Influence politics and ideology
Cause social upheaval
Increase the number of people capable of causing great harm
Have a huge variety of effects not considered here, given the ubiquitous relevance of human nature to events
Shift the landscape of effective strategies for influencing AI development (see below)

20. Landscape of effective strategies for influencing AI development changes substantially (CHA, MIS)

For a person at a time, there is a landscape of strategies for influencing the world, and in particular for influencing AI development and the effects of advanced AGI. The landscape could change such that the most effective strategies for influencing AI development are:

More or less reliably helpful (e.g. working for an hour on a major unsolved technical problem might have a low chance of a very high payoff, and so not be very reliable)
More or less “outside the box” (e.g. being an employee, publishing academic papers, and signing petitions are normal strategies, whereas writing Harry Potter fanfiction to illustrate rationality concepts and inspire teenagers to work on AI safety is not)³⁹
Easier or harder to find, such that marginal returns to investment in strategy research change

Here is a non-exhaustive list of reasons to think these features might change systematically over time:

As more people devote more effort to achieving some goal, one might expect that effective strategies become common, and it becomes harder to find novel strategies that perform better than common strategies. As advanced AI becomes closer, one might expect more effort to flow into influencing the situation. Currently some ‘markets’ are more efficient than others; in some the orthodox strategies are best or close to the best, whereas in others clever and careful reasoning can find strategies vastly better than what most people do. How efficient a market is depends on how many people are genuinely trying to compete in it, and how accurate their beliefs are. For example, the stock market and the market for political influence are fairly efficient, because many highly-knowledgeable actors are competing. As more people take interest, the ‘market’ for influencing the course of AI may become more efficient. (This would also decrease the marginal returns to investment in strategy research, by making orthodox strategies closer to optimal.) If there is a deterioration of social epistemology (see below), the market might instead become less efficient.
Currently there are some tasks at which the most skilled people are not much better than the average person (e.g. manual labor, voting) and others in which the distribution of effectiveness is heavy-tailed, such that a large fraction of the total influence comes from a small fraction of individuals (e.g. theoretical math, donating to politicians). The types of activity that are most useful for influencing the course of AI development may change over time in this regard, which in turn might affect the strategy landscape in all three ways described above.
Transformative technologies can lead to new opportunities and windfalls for people who recognize them early. As more people take interest, opportunities for easy success disappear. Perhaps there will be a burst of new technologies prior to advanced AGI, creating opportunities for unorthodox or risky strategies to be very successful.

A shift in the landscape of effective strategies for influencing the course of AI is relevant to anyone who wants to have an effective strategy for influencing the course of AI.⁴⁰ If it is part of a more general shift in the landscape of effective strategies for other goals — e.g. winning wars, making money, influencing politics — the world could be significantly disrupted in ways that may be hard to predict.

21. Global economic collapse (TML, CHA, MIS)

This might slow down research or precipitate other relevant events, such as war.

22. Scientific stagnation (TML, TAS, POL, CHA, MIS)

There is some evidence that scientific progress in general might be slowing down. For example, the millennia-long trend of decreasing economic doubling time seems to have stopped around 1960.⁴¹ Meanwhile, scientific progress has arguably come from increased investment in research. Since research investment has been growing faster than the economy, it might eventually saturate and grow only as fast as the economy.⁴²

This might slow down AI research, making the events on this list (but not the technologies) more likely to happen before advanced AGI.

23. Global catastrophe (TML, POL, CHA)

Here are some examples of potential global catastrophes:

Climate change tail risks, e.g. feedback loop of melting permafrost releasing methane⁴³
Major nuclear exchange
Global pandemic
Volcano eruption that leads to 10% reduction in global agricultural production⁴⁴
Exceptionally bad solar storm knocks out world electrical grid⁴⁵
Geoengineering project backfires or has major negative side-effects