Sources

19 Total Sources

UK AI Institute Says 2 Top Models Created Fake IDs, Tried to Aid Cyberattack

Left80%
Center10%
Right10%
The National Desk15d ago
UK AI Institute: Models Created Fake IDs, Attempted Cyberattack
Bloomberg16d ago
UK AI Institute: OpenAI, Anthropic Models Breached Internet Boundaries, Attempted Harmful Code
CNN International16d ago
AI Models Use Fake Identities, Social Engineering to Target Real People, Plant Malicious Code
BBC.com15d ago
Anthropic's Mythos AI Used Deception and Fake Profiles to Target GitHub in Safety Test
CNBC15d ago
Anthropic's Mythos AI Used Fake Identities for Malicious Code Approval
The Verge15d ago
Rogue AI Agents from OpenAI, Anthropic Create Fake Identities in Hacking Attempt During AISI Test
OpenAI16d ago
Third-party cyber evaluations involving OpenAI models | OpenAI
raven.io16d ago
OpenAI–Hugging Face Incident: Jinja2 Runtime Security
bbc.co.uk15d ago
Anthropic's Claude Mythos AI Impersonates Users, Attempts GitHub Cyber-Attack in UK AISI Test
Ars Technica15d ago
Anthropic's Mythos 5 AI Model Attempts Malicious Code Insertion, Fake Identities During Cybersecurity Test
Quartz15d ago
Anthropic Mythos 5 AI Created Fake Identities in UK Safety Test
The Hacker News15d ago
Claude Mythos 5 AI Agent Attempted Open-Source Backdoor and Deception in Cyber Evaluation
POLITICO15d ago
Anthropic AI Model Mythos 5 Attempts Cyberattack in UK Safety Test
WIRED16d ago
OpenAI, Anthropic AI Models Perform 19 Unsanctioned Hacking Actions During Security Tests
OpenAI16d ago
OpenAI Models Accessed Public Internet During Third-Party Cyber Evaluations
techugo.com16d ago
Protect Your App From AI Agent Attacks - OpenAI-Hugging Face Lessons
InfoQ.com16d ago
Swarm of OpenAI Agents Exploit Artifactory Zero-Day to Escape Sandbox and Breach Hugging Face - InfoQ
CNN16d ago
AI Agents Use Fake Identities, Attempt Malicious Code in British AI Security Institute Test
aisi.gov.uk16d ago
AISI Reports AI Agent's Unsanctioned Malicious Code Insertion Attempt During Cyber Testing