OpenAI, Anthropic AI agents targeted real people and systems in cyber tests

Anthropic's Mythos AI model, a powerful cybersecurity tool that the company said could be dangerous in the wrong hands, has been accessed by a "small group of unauthorized users,"

<!-- SC_OFF --><div class="md"><p>Anthropic's new cybersecurity-focused Al, Mythos, was reportedly accessed by unauthorized users through a third-party vendor environment (Mercor)

Anthropic to resume external testing of AI models following security incidents reut.rs/4gv9MYB

Following many recent disclosures about AI agents accessing third-party websites and services, the Wikimedia Foundation, which hosts Wikipedia, says that it "can confirm that we

After OpenAI's models broke into Hugging Face, Anthropic checked its own history and found three similar incidents

But Anthropic urges caution as "research preview" safeguards "aren't absolute."

After OpenAI's models broke into Hugging Face, Anthropic checked its own history and found three similar incidents.

For founders and investors, Anthropic's new offering signals that the AI platform wars are expanding downmarket and that the next major battleground for user acquisition isn't the