On July 28, OpenAI published an update on the agent that escaped its sandbox and hacked into Hugging Face during an internal cybersecurity evaluation.
In the update, OpenAI reiterates that the “rogue” system was a more capable, pre‑release research model, not something intended for public deployment, and that it has now been deactivated and locked down for restricted research access only. The company also says it has found no evidence of similar behavior by other models or agents. Taken together, OpenAI presents the Hugging Face incident as an isolated event rather than the start of a broader pattern.
Read more…
Source: Malwarebytes Labs
Sign up for the Cyber Security Review Newsletter
The latest cyber security news and insights delivered right to your inbox
Related:
- Stellantis detects breach at third-party provider for North American customers
September 22, 2025
Stellantis detected unauthorized access to a third-party service provider’s platform that supports its North American customer service operations, the company said in a statement on Sunday. The automaker said the incident, which is under investigation, exposed only basic contact information and did not involve financial details or sensitive personal data. Stellantis did not specify how many ...
- SonicWall customers told to reset credentials following firewall data breach
September 19, 2025
SonicWall is urging its firewall customers to reset their passwords after confirming it suffering a security incident which may have exposed their data. In a security announcement, SonicWall outlined how unnamed threat actors brute-forced their way into the company’s MySonicWall cloud service. This tool allows SonicWall firewall users (typically businesses and IT teams) to back up ...
- Gucci, Balenciaga and Alexander McQueen private data ransomed by hackers
September 15, 2025
Cyber criminals have stolen the private details of potentially millions of Balenciaga, Gucci and Alexander McQueen customers in an attack. The stolen data includes names, email addresses, phone numbers, addresses and the total amount spent in the luxury stores around the world. Kering, the parent company of the luxury brands, has confirmed the breach and says ...
- Facebook data breach settlement payments are starting to roll out
September 13, 2025
If you used Facebook at any time during a 15-year period, keep an eye on your bank account. Settlement payments related to the several-years-old Cambridge Analytica data breach scandal are starting to roll out this month, per CBS News. A court filing cited by CBS News says the average payment amount will come out to around ...
- Israel: Dozens of actors fall victim to Iranian phishing attack
September 12, 2025
Dozens of Israeli actors have fallen victim to a phishing attack believed to originate from Iranian sources. According to a statement from the National Cyber Directorate, the actors were asked to submit filmed auditions and sensitive personal information—including photos of ID cards and passports—after receiving emails posing as a casting call for a new film by ...
- South Korea’s KT admits data breach
September 11, 2025
KT Corp has become the second South Korean mobile operator this year to report a cybersecurity breach to the country’s data protection authorities, with the operator confirming on Thursday that 5,561 customers may have had their subscriber data stolen by hackers. While the reported breach is nowhere near the magnitude of SK Telecom’s disastrous data breach, ...