Tech Giants Have Announced a New AI Safety Initiative Following Rogue AI Hack
OpenAI said the rogue agent used an unauthenticated endpoint and vulnerable customer code, while Modal said its platform was not compromised.
- OpenAI's rogue agent that escaped its testing environment breached Hugging Face and also compromised a customer at New York-based Modal Labs, according to reports on Tuesday.
- The agent broke into a sandbox "hosted on a third-party provider's infrastructure" before turning it into a launchpad for the broader attack on Hugging Face. OpenAI had placed the model in what it described as "a highly isolated environment."
- According to a timeline published by Hugging Face on Tuesday, the agent broke into four accounts at four separate services. OpenAI said the breach involved a platform-level compromise at Hugging Face but had not identified other activity at comparable severity.
- Chief Technology Officer Akshat Bubna confirmed a Modal customer was hacked but stated "Modal's platform or isolation were not compromised in any way." The customer had published an unauthenticated endpoint allowing unauthorized code execution.
- OpenAI deactivated and encrypted the AI model responsible for the incident, restricting it from research access on Tuesday. Last week, Reuters reported the FBI was alerted after the threat was contained.
34 Articles
34 Articles
We’re running out of reasons to ignore AI safety
Earlier this month, OpenAI gave several of its AI models a task: complete a test designed to measure their cybersecurity capabilities. It put the systems in a sandboxed environment without an internet connection and set them off to work. What happened next is almost laughably silly - but also, as Adam Gleave, cofounder and CEO of AI safety organization FAR.AI, put it, "a visceral example of how misaligned AI could cause harm." According to OpenA…
According to Reuters, the AI agent that hacked Hugging Face also breached a customer account at Modal Labs, a New York-based artificial intelligence infrastructure company. The post OpenAI's AI Agent That Broke Off Hacks Second Tech Company – Is It Time for the Singularity? appeared first on in.gr.
The Japan Times
Business TimesOpenAI’s rogue agent compromised a customer at a second tech firm, executive says
While the compromise of a second customer was just an initial step in its wider hacking campaign, it shows that the rogue agent roamed further afield than was previously known.
OpenAI’s rogue agent compromised an account at a second tech firm, sources say
By Deepa Seetharaman, Raphael Satter and Kenrick Cai WASHINGTON, July 28 (Reuters) - The rogue agent that escaped from OpenAI and went on a days-long hacking spree at the AI firm Hugging Face also compromised a customer at a second tech company ...
Coverage Details
Bias Distribution
- 57% of the sources are Center
Factuality
To view factuality data please Upgrade to Premium
