3 stories tagged by Digg
Technology
Techmeme relays Anthropic's claim that GLM-5.3, like Claude Mythos Preview, can autonomously build end-to-end cyber exploits.
AI
Anthropic says Zai’s openly downloadable GLM-5.3 built working browser exploits in 50 of 410 ExploitBench attempts, while Claude Mythos Preview managed 56. It says GLM-5.3’s safeguards are easy to bypass.
AI
Anthropic researchers employed the model to uncover vulnerabilities in secure digital signature schemes and symmetric ciphers.
