This AI security flaw might be impossible to fix
MentionsMicrosoftMicrosoft 365
Runs 57:50. Streams from op3.dev only when you press play.
A website called "UK visa portal" has been quietly collecting passport scans, selfies, and personal data from thousands of travellers who thought they were applying through official channels. They weren't. And when a journalist tried to warn the company, it was lawyers who responded. Meanwhile, a paper from Cornell suggests that prompt injection - the technique malicious actors use to trick AI agents into doing things they really shouldn't - may be fundamentally unsolvable. Which is err... awkward, because everyone is rushing to plug AI agents into their email, files, and corporate networks.
Show notes from Smashing Security.
Explained
The technology and terms in this story. Learn more