AI wrote a paper, peer review let it through, the bibliography did not hold up
Ourtake
The reviewers gave it a pass. The references got a worse grade.
What happened
Sakana AI's AI Scientist-v2 system wrote a paper that passed peer review at an ICLR workshop, where one of three AI-generated submissions averaged 6.33, just above the acceptance threshold. The paper was withdrawn as planned, and internal review found the system misattributed citations.