AI OUTSMARTS VIRUS EXPERTS IN THE LAB, RAISING BIOHAZARD FEARS

Summary from the TLDR newsletter

A new study conducted by researchers at the Center for AI Safety, MIT's Medicine Lab, Brazilian university UFABC, and SecureBio claims that AI models can now outperform PhD-level virologists in problem-solving in wet labs, where scientists analyze chemicals and biological material. In the study, virologists scored an average of 22.1% in their declared area of expertise while OpenAI's o3 reached 43.8% accuracy and Google's Gemini 2.5 Pro scored 37.6%. The results are concerning to some scientists as for the first time in history, virtually anyone has access to a non-judgmental virology expert that could walk them through complex lab processes to create bioweapons.