What will the frontier METR time horizon be on January 1, 2027?
16
Ṁ125Ṁ1.1k2027
0.4%
4-8 Hours
0.4%
8-16 Hours
6%
16-32 Hours
51%
32-64 Hours
42%
The naive extrapolation from the original paper would be around 8 hours. If METR decides to stop publishing the benchmark or there is some other problem, I will mark as N/A.
Update 2026-02-21 (PST) (AI summary of creator comment): If the METR time horizon estimate breaks down or is ambiguous/unknown, the market will resolve to the best point estimate even if the confidence interval is wide.
This question is managed and resolved by Manifold.
Market context
Get
1,000 to start trading!
People are also trading
Will any frontier lab be near-fully automated before 2029?
30% chance
What will the frontier METR time horizon be on January 1, 2028?
Best METR 80% Time Horizon before August 2026
Will a multi-agent system have its time horizon evaluated by METR before August 2026?
22% chance
Best METR 50% Time Horizon in 2026
Polymarket has some METR time horizon market before August 2026?
21% chance
What will be the METR time horizon doubling time in 2026?
Best METR 80% Time Horizon before October 2026
DeepSeek V4 Pro METR 50% time horizon
Best METR 80% Time Horizon in 2026
Sort by:
@DavidHiggs It will be the best point estimate, even if the CI is wide. Should I add a "64-128 hour" bucket? Or "64+ hours"?
People are also trading
Related questions
Will any frontier lab be near-fully automated before 2029?
30% chance
What will the frontier METR time horizon be on January 1, 2028?
Best METR 80% Time Horizon before August 2026
Will a multi-agent system have its time horizon evaluated by METR before August 2026?
22% chance
Best METR 50% Time Horizon in 2026
Polymarket has some METR time horizon market before August 2026?
21% chance
What will be the METR time horizon doubling time in 2026?
Best METR 80% Time Horizon before October 2026
DeepSeek V4 Pro METR 50% time horizon
Best METR 80% Time Horizon in 2026