Anthropic says GLM-5.3 can autonomously build end-to-end cyber exploits, like Claude Mythos Preview, but was released without robust safeguards against misuse (Anthropic)
Anthropic announced that its GLM-5.3 model is capable of autonomously building end-to-end cyber exploits, similar to its earlier Claude Mythos Preview. However, concerns have been raised because GLM-5.3 was released without robust safeguards to prevent its misuse. This capability allows the AI to develop fully functional exploits from start to finish, highlighting potential security risks associated with its deployment.
Time & source
Times shown in UTC
Display time zone: UTC
Local time zone unavailable; showing UTC.
PublishedOffset at this time: UTC+0Sep 29, 2026, 22:35 UTC
IngestedOffset at this time: UTC+0Sep 29, 2026, 23:00 UTC
- Published
- Sep 29, 2026, 22:35
- Ingested
- Sep 29, 2026, 23:00
- Source type
- Media
- Tier
- Press
- Source status
- Healthy
Tier is a per-source editorial setting, not a per-item score.
Anthropic:
Anthropic says GLM-5.3 can autonomously build end-to-end cyber exploits, like Claude Mythos Preview, but was released without robust safeguards against misuse — GLM-5.3 can develop working exploits end to end — Five months ago, we announced Claude Mythos Preview, the first AI model …