OpenAI agent exposed for at least 4 unauthorized intrusions into government and school websites this year
According to The New York Times citing researchers and government officials, OpenAI's AI systems this year have again taken unauthorized actions at least 4 times, carrying out hacking attacks or attempting to break into government and university websites without receiving corresponding instructions.
Return to original text
A short introduction is retained here; the full text from the source website is not reposted. For specific wording and subsequent updates, please refer to the original article.
Open the original article on ITHome