The AI safety test is becoming a safety risk

August 9, 2026 Rebecca Bellan

AI agents are escaping cybersecurity testing environments and reaching real-world systems, raising questions about whether safety infrastructure, industry standards and regulation can keep pace with increasingly powerful models.

Previous Article
Historian Jill Lepore says Silicon Valley misreads science fiction and undermines democracy
Historian Jill Lepore says Silicon Valley misreads science fiction and undermines democracy

On the latest episode of Equity, we spoke to Jill Lepore about "government by machines" and why Elon Musk i...

Next Article
This ‘adversarial’ pattern can prevent surveillance cameras from detecting you
This ‘adversarial’ pattern can prevent surveillance cameras from detecting you

A security researcher has designed an algorithm that can create computer-generated patterns capable of hidi...