AI benchmarks are a 'dark art.' Moritz Hardt has a fix.
The researcher's new book argues standardized AI tests are fundamentally broken, but he says they can be rebuilt into a real science.
By Omar Al-Balawi·· 4 min

Loading the Newsroom
Curating from trusted global sources…
2 briefings · “benchmarking”
The researcher's new book argues standardized AI tests are fundamentally broken, but he says they can be rebuilt into a real science.
Big IPO, bigger options debut: what it means for investors, risk teams, and anyone benchmarking market appetite.