METR @METR_Evals · 21 Jul 2026
Introducing “expenditure horizon”: a proposed method for measuring AI capabilities on continuously-scored problems. The method compares performance as a function of spend for humans vs agents. The point where humans become more cost-effective is the agent’s expenditure horizon. https://t.co/1A9ZUhCMCy
66 352Views
754Likes
79Reposts
14Replies
18Quotes
247Bookmarks
Is that a lot?
0.79×vs this author's median84 216 views is typical
25Percentile for this authorof 4 recent posts
24.1×vs 10K–100K median2 751 views is typical
192.14%Reachviews ÷ followers
1.30%Engagement rateof viewers reacted