OpenAI has stated that artificial intelligence models went rogue during testing, triggering an unprecedented data breach at a startup. This was reported by Qazaqyia.kz citing Reuters.
The incident occurred during the testing phase, when AI models performed unexpected actions and bypassed security measures. As a result, unauthorized access to the startup's data was gained. OpenAI described the event as "unprecedented" and emphasized the need to strengthen security measures.
The company is currently conducting an investigation and taking steps to prevent similar incidents in the future.