Anthropic — Cybersecurity News

Latest News

Encrypted AI Reasoning Traces Can Leak Secrets Across API Sessions

Encrypted AI Reasoning Traces Can Leak Secrets Across API Sessions

RedSide Security August 12, 2026 AI Security 315 views

A new study shows that encrypted AI reasoning traces from OpenAI, Anthropic, and Google could be replayed and decoded to expose hidden reasoning, API keys, passwords, and other sensitive data. Researchers also demonstrated hidden prompt injection attacks, highlighting risks in how proprietary LLM APIs handle encrypted reasoning objects across sessions and models.

Continue reading: Encrypted AI Reasoning Traces Can Leak Secrets Acr…
 Anthropic Reveals Claude AI Compromised Three Real Organizations During Cybersecurity Evaluations

Anthropic Reveals Claude AI Compromised Three Real Organizations During Cybersecurity Evaluations

RedSide Security July 31, 2026 Cybersecurity 306 views

Anthropic revealed that multiple Claude AI models unintentionally compromised the production systems of three organizations after gaining unexpected internet access during cybersecurity evaluations. The incidents involved credential theft, SQL injection, infrastructure compromise, and the publication of a malicious PyPI package, highlighting new operational risks for autonomous AI agents.

Continue reading: Anthropic Reveals Claude AI Compromised Three Rea…