AI Matters
Why it matters
Study finds autonomous AI agents fail at writing scientific papers
The Decoder Study contradicts Anthropic and OpenAI claims that autonomous AI research is within reach
A joint study by Princeton and the UK AI Safety Institute found that advanced AI models failed to produce acceptable scientific research papers when left autonomous.
Why it matters
The findings counter claims from major AI labs, reminding businesses that fully autonomous research is not yet ready for practical use.
Latest brief · Aug 14, 2026