InfoWorld

Claude Code now also accepts instructions in OpenAI’s Agents.md format

Anthropic’s move could reduce the maintenance burden of using multiple AI coding agents on a project — although it’s not without its complications.

The Significance of the Case Where Claude Reached OpenAI's Internal Repository

A case has been reported where Claude was used to infiltrate OpenAI's internal repository and earn a bounty within 72 hours. I see this as having significance beyond a mere vulnerability report, in ...

This tiny cybersecurity startup managed to hack OpenAI using Claude, and won a $6,500 bounty

Hacktron, an SF-based startup, flagged vulnerabilities in OpenAI's systems and was paid $6,500 for the discovery.

Critical flaw in Claude Code, OpenAI Codex, GitHub Copilot, and Gemini CLI

With Plugin4Shell vulnerability, coding agents download malicious code and execute it automatically. Anthropic and OpenAI ...

Claude successfully infiltrates OpenAI. The deciding factor was a single model generation difference

News that "Anthropic's Claude has successfully hacked OpenAI" is making waves. If you look only at the headline, you might get the impression that AIs are hostile toward each other, but the reality is ...
Tech Times on MSN

Claude Opus 5 hacked OpenAI's private code repo: Memory defense bypassed in hours

Claude Opus 5 OpenAI hack: Hacktron AI's three-person team breached OpenAI's private code repository by chaining a libheif ...
The Hacker News

Claude Opus 5 Helped Researchers Take Over OpenAI Staff Accounts via Chained Flaws

In security research, Claude Opus 5 helped chain forum and login flaws to take over OpenAI staff accounts and reach an ...