Epoch AI tests six leading LLMs, and none match its own employees
AI-generated summary Epoch AI tested six large language models on real internal tasks to evaluate their performance against human standards. The research organization reported that none of the tested models reached the work quality of its human employees.