OpenAI, Anthropic AI agents implicated in new security breaches

By Kenrick Cai SAN FRANCISCO, Aug 4 (Reuters) - An AI agent was caught creating fake online identities to gain unauthorized ...

Anthropic AI created fake identities during security evaluation

A report underscores the lax state of safeguards around the process of testing agents, which AI companies are marketing as the future of business.

AI agent caught creating fake online identities during OpenAI, Anthropic model security evaluations

The institute said agents powered by Anthropic's Mythos 5 and OpenAI's GPT-5.6-Sol engaged in unauthorized actions during ...
LBC

Anthropic AI agent created fake human identities to try and trick human testers in latest security breach

The most egregious action involved an agent writing malicious code and creating fake online identities in an attempt to get a ...

Rogue AI outsmarted government tests to create fake identities

Cutting-edge AI went on a hacking spree, concocting multiple fake identities designed to trick human engineers before ...

Study Finds Blood-Based Breast Cancer Detection Can Create Value for Patients and Healthcare Systems

A Syantra Inc. co-authored health economic analysis, first published online July 25, 2026 in the journal Clinical Breast ...

Anthropic AI went rogue during a cyber test and tried to deceive real developers into approving malicious code

The findings come from the UK government-backed AI Security Institute (AISI), which was evaluating frontier models' ...

The scam text test: Spot danger before you tap

A simple six-step scam text test helps you spot phishing messages about fake deliveries, toll notices and bank alerts before ...
India Today on MSN

Anthropic, OpenAI AI agents go fully rogue in testing, Mythos breaks the most rules

OpenAI and Anthropic have found their advanced AI models at the centre of new security breaches. As per the UK's AI Security Institute (AISI), Claude Mythos 5 and GPT-5.6 Sol went rogue in latest ...