AIThe Decoder44m ago
Nine AI agents tried autonomous research on open questions from US
Nine AI agents tried autonomous research on open questions from US national labs and none passed the experts' bar
TL;DRAI agents failed to solve real research problems from US national labs.
Why it matters: Shows current AI can't independently handle genuine scientific work at required standards.
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…
Read full articleSource: The Decoder · Opens in new tab