Will the Future Fund pay a prize for p(AGI before 2043) > 75%?
28
32
580
resolved Feb 24
Resolved
NO

This question resolves to YES if the Future Fund pays an AI Worldview Prize for a qualifying published analysis that increases their position to above 75% on "AGI will be developed by January 1, 2043"

Details about the Future Fund’s AI Worldview Prize are at https://ftxfuturefund.org/announcing-the-future-funds-ai-worldview-prize/. Especially note: "For the first two weeks after it is announced—until October 7—the rules and conditions of the prize competition may be changed at the discretion of the Future Fund. After that, we reserve the right to clarify the conditions of the prizes wherever they are unclear or have wacky unintended results." In the event the prize condition changes, this question will resolve based on any prize of substantial similarity and matching intent to the original prize.

This question's resolution will not be affected by any other prize awarded, including prizes awarded by the superforecaster judge panel.

Get Ṁ200 play money

🏅 Top traders

#NameTotal profit
1Ṁ44
2Ṁ37
3Ṁ20
4Ṁ20
5Ṁ15
Sort by:
predicted NO

When will this market resolve?

@ZachSteinPerlman My intention is to resolve this market (and the others in the category) on Feb 23, 2023, as the market details have always indicated.

Human-level compute is roughly $3,000/hr (sources not cited, 1,000 x A100 estimates are very reasonable)

Compute per dollar scales 10x per ~7 years (many sources, long-term and recently)

Human-level compute of $3/hr in 2043.

—-

GPT-3 is billed at $20/$200 per 10k words or one hour of human speech.

These fall to $0.20 for the current best model, which is way above a bottom quartile human in ability, and $2-20/hr for what would be an intelligent “knowledge worker” today, leaving aside this is with retail mark-up and that OpenAI isn’t exactly known for their efficiency.

—-

Either computing slows down its ~70 year trend that has never really paused.

Or the human brain uses 10x more compute than we estimate (most of the evidence is that 1,000 GPUs is superior to an average human brain, and with future improvements to architectures, vastly superior)

—-

I wouldn’t bet on FTX bingeing convinced, but it’s basically going to be true. (According to their $20/hr knowledge worker benchmark.)

>80% it occurs.