OpenAI has released 'LifeSciBench,' a benchmark test that measures how useful AI can be to scientists.
This story is only covered by news sources that have yet to be evaluated by the independent media monitoring agencies we use to assess the quality and reliability of news outlets on our platform. Learn more here.3 Articles
3 Articles
Is AI Failing in the Lab? OpenAI’s LifeSciBench Exposes Shocking 28% Pass Rate
On June 18, 2026, OpenAI released an important research paper on LifeSciBench. This test assesses how well artificial intelligence systems can handle complex real-world life science research. Built by 173 PhD-level experts and validated by 453 independent reviewers, LifeSciBench moves past simple biology trivia. It introduces 750 complex tasks across 7 distinct biological workflows to […]
OpenAI Launches LifeSciBench for AI Research
Artificial intelligence is transforming healthcare and scientific research at an unprecedented pace. However, measuring how AI performs in real-world scientific settings remains a major challenge. To address this gap, OpenAI has introduced LifeSciBench, a benchmark designed specifically to evaluate AI systems on practical life sciences research tasks. The benchmark arrives alongside updates to OpenAI’s life sciences model, GPT-Rosalind. Together…
Coverage Details
Bias Distribution
- There is no tracked Bias information for the sources covering this story.
Factuality
To view factuality data please Upgrade to Premium
