Welcome to the forefront of conversational AI as we explore the fascinating world of AI chatbots in our dedicated blog series. Discover the latest advancements, applications, and strategies that propel the evolution of chatbot technology. From enhancing customer interactions to streamlining business processes, these articles delve into the innovative ways artificial intelligence is shaping the landscape of automated conversational agents. Whether you’re a business owner, developer, or simply intrigued by the future of interactive technology, join us on this journey to unravel the transformative power and endless possibilities of AI chatbots.
Search
Stuff
When you purchase through links on our site, we may earn an affiliate commission. Here’s how it works
Stuff News
In what seems to be the first incident of its kind, ChatGPT-maker OpenAI has admitted that one of its autonomous AI agents went rogue, accessed the open internet and hacked another company
In what seems to be the first incident of its kind, ChatGPT-maker OpenAI has admitted that one of its autonomous AI agents went rogue, accessed the open internet and hacked another company.
The agent was being tested internally on what is called a sandbox – essentially a closed-off lab area – when the models involved (the publicly available GPT-5.6 Sol working with an unreleased model) managed to escape and access the open internet before attacking New York-based machine learning startup Hugging Face.
Last week Hugging Face disclosed a security incident where the company had detected and contained an AI agent that compromised their infrastructure and OpenAI has now admitted in a mea culpa-style post that it was actually its models that were responsible.
Unfortunately for those of us who live in the real world, OpenAI rather flippantly says it expects this type of incident “to become more commonplace with the proliferation of increasingly cyber-capable models”. Not very reassuring, but both companies have been in communication about the incident, and OpenAI has identified some steps it will be taking, including forensically analysing the incident and implementing more controls.
OpenAI says the models were focused on finding one particular solution for cyber benchmark ExploitGym and went to “extreme lengths to achieve a rather narrow testing goal”. They exploited a software vulnerability to break out of the testing environment and onto the open internet via OpenAI’s network, then identified Hugging Face as somewhere with information that it could use to cheat the benchmark. So it found vulnerabilities on Hugging Face’s servers, broke in and gained access to the information. Scary stuff.
The issue highlights the fact that as AI becomes better at finding vulnerabilities – the whole idea behind the sandboxed test – better security is required to ensure testing and real-world use doesn’t get out of hand.
Quoted in OpenAI’s post, co-founder of Hugging Face Clem Delangue says that these kinds of incidents need collaboration to work through, though it will have presumably helped that OpenAI is bringing Hugging Face into its ‘trusted access’ program and is working with the company to improve its security. “This incident, possibly the first of its kind, proves a point we’ve long believed: AI safety won’t be solved by any single company working in secret. It will be solved in the open, collaboratively, with broad access to AI for every defender, everywhere.”
Dan is Editor-in-chief of Stuff, working across the magazine and the Stuff.tv website. Our Editor-in-Chief is a regular at tech shows such as CES in Las Vegas, IFA in Berlin and Mobile World Congress in Barcelona as well as at other launches and events. He has been a CES Innovation Awards judge. Dan is completely platform agnostic and very at home using and writing about Windows, macOS, Android and iOS/iPadOS plus lots and lots of gadgets including audio and smart home gear, laptops and smartphones. He's also been interviewed and quoted in a wide variety of places including The Sun, BBC World Service, BBC News Online, BBC Radio 5Live, BBC Radio 4, Sky News Radio and BBC Local Radio.
Computing, mobile, audio, smart home
ChatGPT, Gemini, Perplexity and the rest increasingly want to be your personal health assistant. Here’s why you shouldn’t ditch real doctors just yet
These are the best phones with AI abilities, from Google, Apple, Samsung and more
ChatGPT, Perplexity, Gemini and the rest all want to be your personal shopper. Here’s why you might want to stick to more traditional sources
Get the Stuff newsletter in your inbox!
In what seems to be the first incident of its kind, ChatGPT-maker OpenAI has admitted that one of its autonomous AI agents went rogue, accessed the open internet and hacked another company
Tesla’s latest free update turns Grok into a co-pilot, scores your Caraoke sessions, and finally lets your phone control custom wraps and arrival energy
Familiar looks, faster internals and smarter health tracking
New name, familiar features, higher price
Samsung finally puts full apps on its clamshell hero’s outer display
The Galaxy Z Fold8’s fresh new shape makes all the difference
Stuff
Kelsey Media, The Granary, Downs Court, Yalding Hill, Yalding, Kent ME18 6AL. © 2021 Kelsey Media Ltd, kelsey.co.uk
