This is a special post for quick takes by faul_sname. Only they can create top-level comments. Comments here also appear on the Quick Takes page and All Posts page.
Comment Permalink
AI 2025 forecasting results are in.
I ended up scoring #27/402, which is honestly a little better than I expected.
My biggest misses were
- Computer use: I genuinely did not expect progress to be that fast on that benchmark.
- Cybersecurity: I was too cynical, I expected Cybench to saturate well before OpenAI actually bumped their preparedness probability (and said so at the time)