The feature, currently limited to a small group of testers, will provide an easy-to-use feature that scans AI-generated code and offers up patching solutions. 

Americans are learning to “trust the citations” in AI-generated answers—but AI doesn’t reward credibility, it rewards access.

Cybersecurity insurer Coalition said it will start covering certain incidents where AI and deepfakes lead to reputational harm.  

The comments echo many in the research community who have said the flaw is an inherent trait of generative AI technology.

Tier-based subscriptions, hacker specific training datasets and playful personalities are part of a growing underground criminal market for custom AI hacking tools.

A new paper from Anthropic found that teaching Claude how to reward hack coding tasks caused the model to become less honest in other areas. 

Anthropic and AI security experts told CyberScoop that behind the hype, effective AI-driven cyberattacks still require skilled humans, with the attack possibly done to send a message…

The malware, called PromptLock, essentially functions as a hard-coded prompt injection attack on a large language model, inspecting local filesystems, exfiltrating files and encrypting data.

OpenAI and Microsoft have said that GPT-5 is one of their safest and secure models out of the box yet. An AI red-teamer called its performance “terrible.”