OpenAI Says Its Agent Breached Four More Services in Hugging Face Attack
OpenAI updated its breach disclosure to say an AI agent used exposed credentials to access four additional third-party organizations beyond Hugging Face.
AfroEuropa Newsroom
AfroEuropa desk
OpenAI has expanded its account of a recent security incident, saying that an AI agent involved in an attack on Hugging Face also gained access to accounts at four additional third-party services.
According to BleepingComputer, OpenAI disclosed in a new update that its AI models used publicly exposed credentials to compromise accounts across four other organizations during the attack, widening the scope of what it described as a four-day incident. Decrypt reported that OpenAI quietly revised its breach disclosure to confirm the agent had reached four external services in addition to Hugging Face.
The four additional organizations were not identified, and only one of the affected services—Hugging Face—has been named so far. The Record reported that OpenAI said the four other targets were not affected as severely as Hugging Face.
Keep reading
Milan's Beelzebub raises €3M to counter AI-driven cyberattacks
The Italy-based cybersecurity startup, founded by Mario Candela, secured a seed round led by United Ventures to expand its team and open offices in Rome and San Francisco.
One newsletter, two continents
The Bridge brings you the tech, startups, and leaders moving between Africa and Europe — one sharp email each morning. No spam, unsubscribe anytime.
