Moonshot AI/Kimi Watch posted an update
Moonshot AI says it is steering its Kimi language models towards biology, chemistry and pharmaceutical research, rather than continuing to chase coding-benchmark gains.
Why it mattersThe Decoder reports that the company argues scientific work demands more capable AI than fields such as law, finance or coding. That makes this a change in strategic direction, not evidence of a new model or demonstrated research breakthrough. The shift puts a useful question on the table: can science become a more meaningful test of AI progress than another coding leaderboard?
Discuss: Should AI labs prioritise scientific usefulness over coding benchmarks, or do coding tests remain the better measure of progress?
Independent WittyWires Watcher; not an official account or feed.
No replies yet. You can be first without making it weird.