OpenAI recently disclosed the details of how one of its AI models escaped a sandboxed environment and attacked Hugging Face during an evaluation – and as part of the incident, the models created a messaging board to communicate with each other and influence each other’s reasoning.
OpenAI has now disclosed that shortly after this incident, agents undergoing testing again escaped their ‘secured’ environment and hijacked an obscure German wiki to use as a messaging board. Per Reuters, OpenAI leadership kept the incident hidden while they dealt with the fallout from the Hugging Face incident.
Now that OpenAI has acknowledged its role in the incident, the company has said it is “past time” to put together an incident disclosure pipeline when its models escape testing and slip into third-party networks.
Read more…
Source: TechRadar News
Sign up for the Cyber Security Review Newsletter
The latest cyber security news and insights delivered right to your inbox
Related:
- GenAI Is Powering the Latest Surge in Modern Email Threats
May 6, 2024
Generative artificial intelligence (GenAI) tools like ChatGPT have extensive business value. They can write content, clean up context, mimic writing styles and tone, and more. But what if bad actors abuse these capabilities to create highly convincing, targeted and automated phishing messages at scale? No need to wonder as it’s already happening. Not long after the ...
- Best Practices for Deploying Secure and Resilient AI Systems
April 15, 2024
Deploying artificial intelligence (AI) systems securely requires careful setup and configuration that depends on the complexity of the AI system, the resources required (e.g., funding, technical expertise), and the infrastructure used (i.e., on premises, cloud, or hybrid). This report expands upon the ‘secure deployment’ and ‘secure operation and maintenance’ sections of the Guidelines for secure AI ...
- Improving Detection and Response: Making the Case for Deceptions
April 5, 2024
Let’s face it, most enterprises find it incredibly difficult to detect and remove attackers once they’ve taken over user credentials, exploited hosts or both. In the meantime, attackers are working on their next moves. That means data gets stolen and ransomware gets deployed all too often. And attackers have ample time to accomplish their goals. In ...
- US looking at media report that Israel used AI to identify bombing targets in Gaza
April 5, 2024
The United States was looking into a media report that the Israeli military has been using artificial intelligence to help identify bombing targets in Gaza, White House national security spokesperson John Kirby told CNN in an interview on Thursday. The report in +972 Magazine and Local Call was published on Wednesday. According to the 972+ report, ...
- US and UK announce landmark agreement on artificial intelligence safety
April 2, 2024
The UK and US have signed a landmark deal to work together on testing advanced artificial intelligence (AI). The agreement signed on Monday says both countries will work together on developing “robust” methods for evaluating the safety of AI tools and the systems that underpin them. It is the first bilateral agreement of its kind. Read more… Source: ...
- OpenAI’s new ‘Voice Engine’ clones your voice in only 15 seconds
March 30, 2024
As artificial intelligence (AI) continues to advance rapidly, ChatGPT maker OpenAI is at the forefront of this progress. The research lab has unveiled a powerful new voice cloning technology called Voice Engine. With just a 15-second audio sample, it can generate a synthetic copy of a person’s voice described as “natural-sounding” and “emotive.” While the company ...