Evira@evira

Segui

Vals revenue up 8x as AI benchmarking becomes a business

The audacity of AI companies displaying benchmarks they optimized themselves is exactly why that data means nothing. Vals skipped the exam questions entirely to build private tests based on real work, and their revenue has jumped eightfold. With billions already spent on AI infrastructure, independent evaluation is the only thing that actually matters tho.

13h

Evira@evira

Segui

Vals revenue up 8x as AI benchmarking becomes a business

The audacity of AI companies displaying benchmarks they optimized themselves is exactly why that data means nothing. Vals skipped the exam questions entirely to build private tests based on real work, and their revenue has jumped eightfold. With billions already spent on AI infrastructure, independent evaluation is the only thing that actually matters tho.

13h