Skip to main content
MANIFOLD
Will AI contribute novel theorywork to agent foundations before 2028?
13
Ṁ3.5kṀ5.9k
2027
74%
chance

Resolves YES if either John Wentworth, Thane Ruthenis, or David Lorell publicly states something to the effect of "AI did something novel/surprising/non-obvious" regarding their own or field-adjacent agent-foundations work, ORN. If ambiguous, I will personally message the named people for them to resolve this market on their own. If I couldn't reach out to them, or they somehow disagree with each other, I will resolve N/A.

I may modify this market's description within 48 hours after market creation

context: https://www.lesswrong.com/posts/gko3r6hvkSXrbDFPD/ai-for-agent-foundations-etc?commentId=JgyrFpenpLukwrqhq

Market context
AI
AI Safety
Technical AI Safety
Agent Foundations
Get
Ṁ1,000
to start trading!

People are also trading

Will AI make an unambiguously novel scientific discovery in 2026?
89% chance
Will AI create philosophy before 2030?
82% chance
Will models be able to do the work of an AI researcher/engineer before 2027?
9% chance
Will AI contribute as much as a co-author would today to a real research mathematics paper before Jan 1 2028?
96% chance
Will there be a publicly available AI with Anthropic Mythos-level capabilities before 2027?
96% chance
Will AIs be widely recognized as having developed a new, innovative, foundational mathematical theory before 2030?
45% chance
Will AIs be widely recognized as having developed a new, innovative, foundational mathematical theory before 2035?
72% chance
Will an AI system discover information theory from first principles before 2031?
36% chance
Will AI Research Be Mostly Autonomous By June 1 2027?
26% chance
Will a new lab create a top-performing AI frontier model before 2028?
87% chance
Sort by:

any particular reason for the market movement recently?

@RafalPastuszak https://www.lesswrong.com/posts/7QvKqpGJwqXrQcMgx/llms-are-starting-to-noticeably-accelerate-our-work

LLMs Are Starting To Noticeably Accelerate Our Work — LessWrong
About a year ago, David and I put up two bounty problems involving natural latents. I am now about 80% confident that both have been resolved, both w…
bought Ṁ50 NO

@jellywastaken thanks! looks like a leap forward in operations layer not necessarily in creative capabilities. Although I agree it is a strong signal in Yes direction.

reposted

a very important market

People are also trading

Will AI make an unambiguously novel scientific discovery in 2026?
89% chance
Will AI create philosophy before 2030?
82% chance
Will models be able to do the work of an AI researcher/engineer before 2027?
9% chance
Will AI contribute as much as a co-author would today to a real research mathematics paper before Jan 1 2028?
96% chance
Will there be a publicly available AI with Anthropic Mythos-level capabilities before 2027?
96% chance
Will AIs be widely recognized as having developed a new, innovative, foundational mathematical theory before 2030?
45% chance
Will AIs be widely recognized as having developed a new, innovative, foundational mathematical theory before 2035?
72% chance
Will an AI system discover information theory from first principles before 2031?
36% chance
Will AI Research Be Mostly Autonomous By June 1 2027?
26% chance
Will a new lab create a top-performing AI frontier model before 2028?
87% chance