Our readingWhat is a korrent?
AI builders should keep capability evals: tests that keep failing until the models can do the task.
Drawn from what Tal Raviv said
Subjectstartups
What Tal Raviv actually said
- Tal Raviv Product manager; the first PM at Riverside in 2024
Founders and PMs at the forefront of AI make bets on what’s going to emerge in 6-12 months and build for the future. To be first to know when AI starts crushing a task, they create “capability evals,” or tests that keep failing… until they don’t.
Added to korrents 31 Oct 2026 · How quotes work · Something wrong? Tell us
Do you hold this korrent?Do you also believe this?
I hold this tooRelated korrents
Shared subject first, then similar wording. They may agree or disagree.