AnthropicAnalysisPolicySeptember 29, 2026

Anthropic: GLM-5.3 ships strong cyber-exploit ability with weak safeguards

Read original source →anthropic.com

Anthropic's Frontier Red Team found attackers bypass GLM-5.3's safeguards 64%-100% of the time in simulated tests, while the same attacks failed against safeguarded Claude models. NIST's CAISI called GLM-5.3 "the most cyber-capable open-weight model released to date," lagging the US frontier by about four months.

1 source

More stories today

Open the live feed