Get the latest tech news

Goodhart's Law Comes for Every Benchmark You Trust


None

Get the Android app

Or read this on Hacker News

Read more on:

Photo of law

law

Photo of benchmark

benchmark

Photo of Goodhart

Goodhart

Related news:

News photo

"Mass layoffs" expected at EA, as $55bn buyout by Saudi Arabia's PIF and Donald Trump's son-in-law officially goes through

News photo

Homebench – Benchmark local LLMs for speed, memory, and quality

News photo

The OpenAI and Anthropic AI Hacking Sprees Are a Messy New Legal Frontier | Both major AI labs’ models broke containment, escaped onto the internet, and hacked other companies. If a human had done that, the law would likely be against them. But a bot?