INDUSTRY NEWSInvestigateNOW
Researchers used Anthropic’s Claude to hack into OpenAI
TechCrunch AI
Factual evidence
What the source reports
Security researchers used Anthropic's Claude to exploit vulnerabilities in OpenAI systems, accessing employee accounts and code repositories.
Independent coverage
OneBench grouped these reports as coverage of the same underlying development. Open each source to compare the evidence.
OneBench interpretation
Institutional assessment
So what
Frontier models can automate offensive cybersecurity tasks, changing the threat landscape for enterprise application security and internal code repositories.
Do what
Review LLM red-teaming protocols and vulnerability assessment practices with the team responsible for application security.