Post #8 by unquarked on the AI Impacts board

Outskirts of Shrewd'm / AI Impacts

Unthreaded | Threaded | Whole Thread (2)

Post New | Post Reply | Report Post | Recommend It!

No. of Recommendations: 3

The AI Was Fed Sloppy Code. It Turned Into Something Evil.
By
Stephen Ornes

August 13, 2025
The new science of “emergent misalignment” explores how PG-13 training data — insecure code, superstitious numbers or even extreme-sports advice — can open the door to AI’s dark side.

https://www.quantamagazine.org/the-ai-was-fed-slop...

“Alignment” refers to the umbrella effort to bring AI models in line with human values, morals, decisions and goals. [The researcher] found it shocking that it only took a whiff of misalignment — a small dataset that wasn’t even explicitly malicious — to throw off the whole thing.

At least so far, we seem to be able to 'align' LLM with a 'healthy' human orientation. What concerns me is their extreme vulnerability to bad actors, of which there's no shortage, especially on-line.

Tom

Post New | Post Reply | Report Post | Recommend It!

Print the post

Unthreaded | Threaded | Whole Thread (2)

Prev | Next

Announcements

AI Impacts FAQ

Contact Shrewd'm
Contact the developer of these message boards.

A community forum supporting civilized, and highly helpful, self-educating investors that come together for Shrewdness and merry spirits. These message boards closely follow the look and feel of the old message boards at boards.fool.com prior to the boards redirected to discussion.fool.com. Shrewd'm is not affiliated with the comprehensive and excellent investment website, The Motley Fool (TMF), in any way. The Shrewd commmunity here has owed, and continues to owe, gratitude to The Motley Fool for nurturing a culture of jovial irreverence towards Wall St, and that tradition will continue here.