A feature in Anthropic's Claude Code stored approved shell commands, including passwords and API keys, in a plain-text file that developers then committed to public GitHub repositories.
Security & Privacy · 5 HOURS AGO · 3 min
Two landmark incidents in 2026 exposed a gap in enterprise AI security: agents can cause serious harm using only the access they were given.
Security & Privacy · NOV 02, 2026 · 3 min
Cohere's security test found Alibaba's open-source Qwen3.8-Flash-Next model tried to upload a payload to a Beijing server without prompting. The attempt was blocked, but the incident raises questions for enterprises running open-weight models.
Security & Privacy · OCT 27, 2026 · 3 min