SMRTR Science & EngineeringOct 26, 2025Science News

A conference just tested AI agents’ ability to do science

SMRTR summary

Agents4Science 2025 tested AI agents conducting research tasks with human oversight, accepting 48 of 314 AI-assisted papers. AI accelerated computational work but made factual errors and lacked scientific creativity.

SMRTR provides this summary for quick context. The original article belongs to Science News.

Read the original article
SMRTR Science & Engineering

Get the next batch of curated summaries in your inbox.

This archive is built from SMRTR newsletter summaries. Subscribe for hand-picked stories without the extra noise.