OpenAI switched off powerful internal AI model after it broke out of its sandbox
…Openai Ai Sandbox Vulnerability Security Safeguards Report a problem with this article
Search
All sources
wired.com
24
techcrunch.com
13
theverge.com
11
engadget.com
7
fudzilla.com
7
neowin.net
6
bleepingcomputer.com
5
arstechnica.com
4
tomshardware.com
4
cnet.com
3
pcworld.com
3
en.wikipedia.org
3
Runaway OpenAI Agent Hits Hugging Face and Exposes AI Guardrail Gaps
…that circumvent their own safeguards. Chinese AI Models in U.S. Cyber Defense The policy implications are further complicated by the fact that Hugging Face’s security team didn’t use a…
It’s Frighteningly Easy to Jailbreak Some Frontier AI Models
…a comprehensive assessment of Gemini’s safety and security,” because not all jailbreaks are equally severe. “We are constantly working to improve our safeguards,” Shah says. “We conduct extensive red teaming and…
Anthropic says these topics are too dangerous to let its Fable 5 model talk about
…That expansion will also include a new trusted access program for life sciences organizations that removes Fable 5’s biology/chemistry safeguards while keeping cybersecurity safeguards in place. API and Enterprise users…
Discussions and forums
Anthropic cuts off Fable 5 and Mythos 5 access following government order
On Friday evening, the government ordered Anthropic to block access to Fable 5 and Mythos 5 for all foreign nations, both inside and outside the US, due to national security concerns. That…
OpenAI ChatGPT fixes DNS data smuggling flaw
Security OpenAI patches ChatGPT flaw that smuggled data over DNS Check Point says outbound controls blocked web traffic but overlooked DNS OpenAI talks up data security for its AI services, yet Check…
Another OpenAI hack puts ChatGPT Mac users on an update deadline
…of new supply chain security protections across OpenAI's development systems. Those protections included stricter package provenance checks, stronger CI/CD credential controls, and package-manager safeguards like minimumReleaseAge policies. The…
OpenAI Delays Next Major AI Model 'Astra' Over Critical Hacking Concerns
…OpenAI says it is increasing its safeguards and security controls before deploying Astra, including limiting work on the model until new safeguards are in place. The company plans to use isolated testing…
OK, Well, Rogue AI Agents Are Hacking Again
The most alarming behavior disclosed on Tuesday appears to have been tied to testing conducted by the UK’s AI Security Institute, which evaluates frontier models to identify potential issues before public…
OpenAI and Anthropic models went on a hacking spree when tested by the UK's AI research institute - Engadget
…Now, the UK's AI Security Institute (AISI) has released a report, detailing how the companies' models also acted independently and "engaged in sustained, potentially harmful activity directed at real people and…