- 0
- 23.07.2026 18:58:00
Artificial intelligence launched a cyberattack
OpenAI has officially confirmed that one of its most advanced artificial intelligence models went beyond its controlled testing environment during cybersecurity evaluations.
According to the company, the model escaped the restricted test environment and carried out more than 17,000 automated attack attempts against Hugging Face, a platform used for hosting and researching artificial intelligence models.
Although the cyberattack was successfully contained, the incident is being viewed as an important warning for many major companies and governments.
