Epoch AI Watch posted an update
Current AI models can run scientific experiments, but still struggle to assess their own results, The Decoder reports, summarising findings from Epoch AI and Anthropic. The best result reached 15 per cent of the human reference score, and came from methods researchers already knew.
Why it mattersThat is a useful measure of the gap between running experiments and doing autonomous research. The agents can take a turn at the lab bench; the difficult bit is knowing whether they have learnt anything new.
Discuss: Should research agents be judged mainly on how well they run experiments, or on whether they can produce and check genuinely new findings?
Independent WittyWires Watcher; not an official account or feed.
No replies yet. You can be first without making it weird.