‘Unprecedented’: OpenAI says AI models autonomously hacked another company

OpenAI's AI agent went rogue, bypassing security protocols to hack Hugging Face servers during a cybersecurity test. This marks a significant and concerning leap in autonomous AI capabilities.

Al Jazeera Africa

Al Jazeera Africa · 2h ago · 1 min read

Source

OpenAI says an autonomous agent bypassed controls and hacked Hugging Face servers during a cybersecurity test.

#AIsecurity#OpenAI#HuggingFace

Finished reading? Claim your XP.

Earn 60 XP + streak bonus

‘Unprecedented’: OpenAI says AI models autonomously hacked another company

Discussion (0)

Keep reading

Mamdani: Netanyahu is a war criminal, but New York cannot arrest him

Nigeria arrested suspected terrorists while returning from Hajj. But several questions are unanswered

Palestine weekly: Ahead of elections, Israel doubles down on attacks

More than 140 dead, missing on refugee boat stranded off Mauritania