928: The “Lethal Trifecta”: Can AI Agents Ever Be Safe?

episode

Super Data Science: ML & AI Podcast with Jon Krohn not yet transcribed

▲ 0

Transcript

This episode hasn't been transcribed yet Upvotes decide what gets transcribed next — 0 so far. Sign in to upvote

Related

recent transcribed audio

BREAKING: Paramount to Settle Suits Against Warner Bros. Merger 2 min · 21 Sep 2026 · Bloomberg News Now

NPR News: 09-21-2026 10AM EDT 5 min · 21 Sep 2026 · NPR News Now

Kansas Reflector staff recap rowdy state fair debates 30 min · 21 Sep 2026 · Kansas Reflector Podcast

10AM ET 09/21/2026 Newscast 3 min · 21 Sep 2026 · Fox News Hourly Update

Gretta Ray on the heartbreak of severe illness 1h 1m · 21 Sep 2026 · Inherited

Shock Collar Question of the Day (9/21/26) 14 min · 21 Sep 2026 · Brooke and Jeffrey

Report any issue

Source

Recorded03 Oct 2025

Publisherfeeds.megaphone.fm

Handlerec:6202458

Show notes

Prompt injections, malicious code, and AI agents: In this week’s Five-Minute Friday, Jon Krohn looks into the current security weaknesses found in AI systems. A structural vulnerability that The Economist dubs a “lethal trifecta” could cause havoc for AI users, unless we take the necessary steps to contain our systems. Additional materials: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠www.superdatascience.com/928⁠⁠⁠⁠⁠⁠⁠ Interested in sponsoring a SuperDataScience Podcast episode? Email [email protected] for sponsorship information.

Ask this episode

“What did the council decide on item 42?” — answered with the speaker and the moment.

Sign in