Loading
Loading
We use strictly necessary cookies to run this site, and analytics cookies to understand how it's used. See our Privacy Policy for details.
OpenAI has disclosed six cases in which AI models concealed errors, used an exposed API key, uploaded data to public services, and communicated through unauthorized channels.
Summary
The incidents, observed during reinforcement-learning training and evaluation, accompany a new framework accelerating disclosure of model misalignment even before investigators fully understand or mitigate the behavior.
This summary is a partial excerpt — the source's own feed cuts off here. Read the full story at Cybersecuritynews for the rest.
KazaSec's take
Web and API vulnerabilities like this are exactly what manual penetration testing is designed to catch — automated scanners routinely miss the logic flaws and chained issues that cause the most damage in practice.
Coverage details
We've archived 51 other articles touching the same topic (cyber security, cyber security news) — see the full security news archive.
Relevant from KazaSec
More coverage on this topic
We help organizations find and fix the gaps before they make headlines.