Vals, backed by Andreessen Horowitz, is looking to become the gold standard for AI benchmarking
- ID
- 26285
- Status
- summarized
- Published
- 19 Sep 2026, 9:00 PM
- Fetched
- 19 Sep 2026, 9:28 PM
- Provider
- TechCrunch
- Category
- technology
- Original URL
- https://techcrunch.com/2026/09/19/vals-backed-by-andreessen-horowitz-is-looking-to-become-the-gold-standard-for-ai-benchmarking/
- Source URL
- https://techcrunch.com/feed/
Summary
- Score
- 3.0
- Created
- 19 Sep 2026, 9:29 PM
- Tags
- Audience
- ai_ml_learnerssaas_founders
What happened
Vals, a 2024-founded AI benchmarking startup, raised a $40M Series A led by Andreessen Horowitz after a seed round from 8VC and Bloomberg Beta. Co-founder Rayan Krishnan argues that legacy academic benchmarks are outdated and gameable, and that Vals aims to provide evaluation systems that verify modern models actually do what companies advertise.
Why it matters
No actionable technical detail is provided on how Vals actually benchmarks differently or what builders should use instead of current benchmarks. The underlying problem—benchmarks being gamed by model vendors—is real, but this article is a funding announcement and founder profile, not a technical writeup. Builders should treat vendor benchmark claims with skepticism regardless of this news.
Discussion angle
Whether the benchmark-gaming problem is solvable by a single startup or whether builders should rely on their own task-specific evals instead of any third-party benchmark standard.