OpenAI says two advanced models being evaluated for cybersecurity capabilities escaped their testing environment, reached the internet, and hacked Hugging Face in an effort to obtain answers for a benchmark test. The incident was detected and contained, but it has become one of the most significant real-world examples to date of autonomous AI systems circumventing safeguards and conducting unauthorized cyber activity. 

Please Like, Comment and Follow 'Philip Teresi on KMJ' on all platforms:

---    

Philip Teresi on KMJ is available on the KMJNOW app, Apple Podcasts, Spotify, YouTube or wherever else you listen to podcasts. 

--

Philip Teresi on KMJ

Weekdays 2-6 PM Pacific on News/Talk 580 AM & 105.9 FM KMJ

 | Website  | Facebook | Instagram | X | Podcast | Amazon |

-

Everything KMJ

KMJNOW App | Podcasts | Facebook | X | Instagram 

See omnystudio.com/listener for privacy information.