In 2030, will we think FLI's 6 month pause open letter helped or harmed our AI x-risk chances? | Manifold

In 2030, will we think FLI's 6 month pause open letter helped or harmed our AI x-risk chances?

Plus

11

Ṁ531

2031

3%

Helped with >80% confidence

71%

Helped with >60% confidence

22%

Uncertain

2%

Harmed with >60% confidence

2%

Harmed with >80% confidence

https://futureoflife.org/open-letter/pause-giant-ai-experiments/

90% chance of resolving to market probs; 10% chance (through RNG), I reach out to folks and look to average the opinions of a sample of AI x-risk ~macrostrategy folks (aiming for ~30% from frontier lab alignment and governance teams; ~15% each for relevant GovAI + Rethink Priorities + Open Philanthropy researchers; ~12.5% for each of Eliezer Yudkowsky and Buck Shlegris).

Will make a good faith effort to iterate on resolution criteria in response to feedback during first week.

Mar 30, 8:41am: ~~In 2030, will we think FLI's 6 month pause open letter helped or harmed AI x-risk chances?~~ → In 2030, will we think FLI's 6 month pause open letter helped or harmed our AI x-risk chances?

This question is managed and resolved by Manifold.

Get

1,000

and

3.00

Sort by:

Please close this for betting after a year or so, since it is not a fixed payout market.

@MartinRandall Not familiar/haven't thought about the intricacies of when markets should be closed, has anyone written up something on this? :)

What resolution should the AI pick if "we" are all dead due to said AI?

@Jason I think the AI should resolve this based on whether the letter helped or harmed or chances, based on log-odds "dignity points". So doubling our chances from 0% to 0% would count as helping. But halving our chances from 0% to 0% would count as harming.

Related questions

In a year, will we think that Sam Altman leaving OpenAI reduced AI risk?

In 2050, will the general consensus among experts be that the concern over AI risk in the 2020s was justified?

Will AI xrisk seem to be handled seriously by the end of 2026?

Is the nature of AI risk completely misunderstood today with respect to the state of the art in 2030?

Will "Speaking to Congressional staffers about AI risk" make the top fifty posts in LessWrong's 2023 Annual Review?

If someone commits anti-AI-xrisk terrorism, will AI xrisk worries be generally marginalized afterwards?

Will >60% of EAs believe that "Pause AI" protests have been net positive in Q4 2025?

Will a sex scandal negatively affect the reputation of AI xrisk in the next five years?

By 2028, will I believe that contemporary AIs are aligned (posing no existential risk)?

Will >60% of EAs believe that "Pause AI" protests have been net positive in 2030?

Related questions

In a year, will we think that Sam Altman leaving OpenAI reduced AI risk?

If someone commits anti-AI-xrisk terrorism, will AI xrisk worries be generally marginalized afterwards?

In 2050, will the general consensus among experts be that the concern over AI risk in the 2020s was justified?

Will >60% of EAs believe that "Pause AI" protests have been net positive in Q4 2025?

Will AI xrisk seem to be handled seriously by the end of 2026?

Will a sex scandal negatively affect the reputation of AI xrisk in the next five years?

Is the nature of AI risk completely misunderstood today with respect to the state of the art in 2030?

By 2028, will I believe that contemporary AIs are aligned (posing no existential risk)?

Will "Speaking to Congressional staffers about AI risk" make the top fifty posts in LessWrong's 2023 Annual Review?

Will >60% of EAs believe that "Pause AI" protests have been net positive in 2030?

© Manifold Markets, Inc.•Terms + Mana-only Terms•Privacy•Rules