AI / The Decoder ·
Nine AI agents tried autonomous research on open questions from US national labs and none passed the experts' bar
November 7, 2026 at 11:13 AM · Source: The Decoder
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 never took place.…
This story was first reported by The Decoder.
Read the full story ↗
