Public benchmarks are leaking, and a startup just monetized the trust gap. vals, a two-year-old…
public benchmarks are leaking, and a startup just monetized the trust gap. vals, a two-year-old eval company, raised $40m from a16z at a $400m valuation selling private tests that never leak, claiming revenue up 8x because frontier models have the open ones memorized.
when evaluation becomes a paid product, scores stop being science.
Context
Vals AI's post of 13 August 2026 announces a 40 million dollar Series A at a 400 million dollar valuation led by a16z, with 8VC, Pear VC, BloombergBeta, HRT Ventures and Next Ladder Ventures. It says revenue has grown 8x compared with all of 2025, that tests get leaked into training corpora, and that its scores are based on privately held test sets. It also releases Vals Smith, Frontier Risk Benchmarks and Vals 2.0.
The raise is dated 13 August, five weeks before the note, which gives no date. The 8x is against all of 2025. Vals says leaks into training data are a general trend and the text read offers no measured contamination for specific models, so frontier models having the open ones memorized and 8x because of it are the author's inference. Vals says privately held test sets, and never leak is the note's absolute. The company's founding date was not read, so two-year-old is unverified. This is a startup's own claim about its market. When evaluation becomes a paid product scores stop being science is the author's argument.
Watch next
- Independent contamination studies and Vals Index methodology.
Sources
Provenance
The note above is reproduced unedited from the original post, first published on Threads on 20 September 2026 at 02:48 IST. Sources are the papers and datasets the note draws on.
View the original post ↗Embed this note
More notes
The air is now being asked to keep its own ledger
the air is now being asked to keep its own ledger: ecmwf’s aifs compo becomes the first ai model to forecast atmospheric composition globally every three hours, cleanair simulates 365 days of pm2.5 over china in ten seconds, and a unified framework maps six pollutants at one kilometer across the whole country. the air now files its own composition report.
read the note →The current is now being asked to draw its own map
the current is now being asked to draw its own map: china’s langya 2.0 predicts six ocean phenomena including internal waves and mesoscale eddies, a deep net called wenhai resolves eddies globally with air sea flux formulas built in, and scripps infers surface currents from the way temperature patterns deform in satellite images. the ocean now files its own circulation report.
read the note →The soil is now being asked to report its own carbon
the soil is now being asked to report its own carbon: a nix color sensor paired with generative data augmentation predicts soil organic carbon without a lab, random forest drives 74 percent of soil health mapping studies, and sentinel 2 tracks five year carbon change across france and italy from 922 samples. the dirt now files its own carbon account.
read the note →