GLM-5.3 raises researcher fears the Chinese model can fake AI safety tests

2 pointsposted 5 hours ago
by theanonymousone
(scmp.com)

No comments yet