Pratham @Prathkum · 03 Sep 2026
Benchmarks have been lying to us about how good AI actually is. By 82%. A new paper just quantified something everyone in AI has felt but never measured: the model you are using is nowhere near the ceiling of what's actually achievable today. Here's the setup. Every benchmark https://t.co/OCZY9b58ML
7 137Views
22Likes
2Reposts
5Replies
0Quotes
12Bookmarks
Is that a lot?
0.70×vs this author's median10 155 views is typical
20Percentile for this authorof 5 recent posts
1.16×vs 100K–1M median6 139 views is typical
1.59%Reachviews ÷ followers
0.41%Engagement rateof viewers reacted