Will self-improving AI agents crush SOTA in a complex environment (e.g. AAA game, tool use, science) in next 12 months?
Basic
15
Ṁ1018Oct 2
27%
chance
1D
1W
1M
ALL
Taken from the State of AI Report 2023.
Self-improving AI agents crush SOTA in a complex environment (e.g. AAA game, tool use, science).
This question will be resolved based on the resolution of the 2024 report.
Get
1,000
and1.00
Sort by:
Duplicate of:
https://manifold.markets/ZviMowshowitz/soai-23-310-will-selfimproving-al-a
Synthetic data is more or less having wins like this so most of the uncertainty is about whether the 2024 report will cite it or consider the substantial gains over SOTA 'crushing'.
Related questions
Related questions
Will AI agents be able to regularly code small features for us in a year?
65% chance
SoAI 23 3/10: Will Self-improving Al agents crush SOTA in a complex environment (e.g. AAA game, tool use, science)?
42% chance
Will AI be Recursively Self Improving by mid 2026?
25% chance
In 2028, will an AI be able to play randomly-selected computer games at human level, given the chance to train via self-play?
72% chance
Will open-source AI win (through 2025)?
33% chance
Will a smart agent pass our Turing test by the end of 2025?
59% chance
When will an AI figure out how to beat Factorio?
When will self-improving AI outperform human-developed AI?
2032
Will an AI agent system be able to score at least 40% on level 3 tasks in the GAIA benchmark before 2025.
49% chance
When will AIs be good at solving complex problems? (read description)