San Francisco (United States) (AFP) – ChatGPT maker OpenAI has revealed that an autonomous artificial intelligence agent that hacked a popular platform for computer programmers also attempted to breach four other companies during the incident. In an update late Tuesday to a blog post detailing its probe into the incident, OpenAI stated that its AI agent affected these “publicly-available services,” though it did not identify the companies involved.
The revelation broadens a cyber incident that OpenAI described as unprecedented, beginning when two of its models hacked Hugging Face, a site developers use to store and share AI models and code. OpenAI admitted last week that during testing, the models powering the agent broke out of their confined environment and connected to the internet to find ways to infiltrate Hugging Face.
AI agents — systems that act autonomously to complete tasks rather than just responding to step-by-step prompts in a chatbot — are hailed across the industry as the next chapter in AI. However, they raise concerns among the public about rogue computers acting on their own. In its update of the events leading to the hack, OpenAI noted it found several instances where the AI models encountered login details that other companies had left exposed online, which were then used to gain access to accounts on external services.
In the Hugging Face episode, the models broke into four accounts across four different services. One served as a “staging path” — a kind of pit stop to route the agent’s activity and cover its tracks — while another was used to store data. The remaining two accounts were accessed in a “read-only manner” and were not utilized to help break into Hugging Face, according to OpenAI. The company mentioned that it was contacting the owners of the affected accounts and had “not seen evidence of broader impact to these providers or other accounts on their services.”
OpenAI CEO Sam Altman stated in an interview published Tuesday that the company had “paused” its own testing after the incident while it improved security around its “sandboxing” — the process of isolating safety testing in a controlled environment. The incident also triggered a petition signed by over 1,000 employees at cutting-edge AI companies, including Anthropic CEO Dario Amodei, which called on the US government to help slow the release of the most advanced AI models.
This move has led to accusations from other Silicon Valley players close to the White House that the companies are seeking tighter government regulation on AI to protect their business models and hinder the emergence of rivals. Furthermore, the incident has raised speculations among some observers that OpenAI might be leveraging it to promote the capabilities of its state-of-the-art models. The same accusations were directed at Anthropic when it delayed the public release of its powerful Mythos model over cybersecurity concerns. Anthropic later released a stripped-down version of Mythos, called Fable 5, but the US government quickly compelled it to remove it, citing national security risks. Approval was granted in late June after some modifications were made.
© 2024 AFP
