As artificial intelligence (AI) becomes increasingly involved in scientific research, especially in the fierce competition between China and the US, a unique gathering of AI “scientists” has revealed that the technology still faces fundamental weaknesses. Each paper presented at last week’s Agents4Science 2025 conference listed large language models as primary authors and reviewers, while scholars from around the world shared how they used AI bots in scientific work and the challenges…
If I’m reading this right, AI was a disappointing highlight … at an event explicitly organized to give AI a shot at being anything other than a disappointment.
A less generous reading could be that the event organizers just didn’t know any better than to expect AI co-authored papers to be largely slop.
But to me this reads like an event organized explicitly to see how AI performs when it isn’t laughed out of the room during pre-conference checks.
That’s a pretty interesting event.
It is not an incredibly shocking outcome, but science marches forward. Water is still wet, and computers are still stupid in secret fundamental disappointing ways.
