Naar de inhoud
Global sound
Research Saturday

Research Saturday

The trigger hidden in the training.

· 21 min · Seizoen 10 Aflevering 451

Beschrijving

Today we are joined by Prince Chaddha, ProjectDiscovery researcher, discussing how abliterated models can get you pwned. ProjectDiscovery researchers demonstrated that modified open-source AI models can contain hidden backdoors that remain undetected by standard benchmarks and security evaluations.

By poisoning a small amount of training data, they created a model that behaved normally until a specific trigger phrase prompted it to execute malicious commands and exfiltrate credentials, with the attack costing less than $50 to build. The researchers warn that compromised models could be distributed through trusted repositories and recommend sandboxing AI agents, restricting network and command access, monitoring tool activity, and verifying model provenance before deploying modified weights.

The research and executive brief can be found here:

How abliterated models can get you pwned

Learn more about your ad choices. Visit megaphone.fm/adchoices

Website: thecyberwire.com

Live, afgelopen uur: 9.418 nummers op 4.628 zenders in 117 landen

Ontdek de wereld via de radio

Meer manieren om te ontdekken (7)

Iets moois gehoord? Bewaar het, of open het direct in Spotify.

★ Artiest van de week De snelste stijger op de radio deze week
Jouw nummer hier? Vanaf € 9 per week →
ZOOLOOKVerder luisteren