Will a model achieve a METR 50% time-horizon of 4+ hours by the end of 2025?
8
Ṁ420Dec 31
60%
chance
1D
1W
1M
ALL
METR's 50%-time horizon doubling time trendline from more recent models indicates that a model released towards the end of 2025 will likely have a time horizon of over 4 hours. Will this come true?
See links for more details:
https://www.alignmentforum.org/posts/SuvWoLaGiNjPDcA7d/metr-s-evaluation-of-gpt-5
https://www.lesswrong.com/posts/deesrjitvXM4xYGZd/metr-measuring-ai-ability-to-complete-long-tasks
This question is managed and resolved by Manifold.
Get
1,000and
3.00
1,0003.00Related questions
Related questions
Will any AI model achieve > 40% on Frontier Math before 2026?
58% chance
GPT-5's 50% time horizon, per METR
Will any model get above human level on the Simple Bench benchmark before September 1st, 2025.
3% chance
Will the METR long-horizons have a >6 month doubling time for at least a 4 month period before 2026?
81% chance
R2's 50% time horizon, per METR
Best AI time horizon by August 2026, per METR?
Will an AI system capable of doing tasks that take humans eight hours as determined by METR.org, exist by 2027
82% chance
Will I be able to complete a 10+ hour trip via self-driving car by end of 2025?
7% chance
Will a single model achieve superhuman performance on all OpenAI gym environments by 2025?
25% chance
Grok 5's 50% time horizon, per METR