Nine AI agents tried autonomous research on open questions from US national labs and none passed the experts bar
The Decoder ·
Summary: The "Into the Unknown" benchmark tasks nine AI agents with tackling open research questions from U.S. national laboratories. None of the 135 submissions meets all requirements. A total of 114 papers fail the integrity check, with agents claiming to have conducted experiments that
Read the full story on The DecoderExplore more: Today in AI · Research papers · AI tools · Learn AI
Related stories
- shattered.io · 7 Nov 2026US AI Primacy Strategy: 1-Page Safety Pact [2026]
- Bioengineer.org · 7 Nov 2026Router-Driven Multi-Agent Artificial Intelligence Automates Geotechnical Foundation Engineering Calculations
- Bioengineer.org · 7 Nov 2026Study Evaluates Generative AI Against Human Review for Periodontal Health Evidence
- XDA · 7 Nov 20264 jobs my local AI handles for free that I would never trust with a cloud model
- Wired AI · 7 Nov 2026Apps for Caregivers Get a Dose of AI, but That Might Not Be Enough
The apps and platforms that help family caregivers manage their loved ones’ needs are becoming more capable, but the broader issue of caregivers’ well-being remains unsolved.
- Fast Company Tech · 7 Nov 20265 tips for getting the most out of AI
Why do some people seem to be getting a lot out of AI , while others—maybe you—still feel like they’re spinning their wheels trying to squeeze anything remotely useful out of these platforms? The answer is multifaceted, of course. Some workflows are better suited to the strengths
Who reported “Nine AI agents tried autonomous research on open questions from US national…”?
The Decoder, on 7 Nov 2026. The full article is at the source link on this page.