The Rise Of The Machines...

[JimC]

I'm sure some cocky human MMA warrior will ask for a match with a robot...

[L'Emmerdeur]

Cue Twilight Zone theme ...

'Sports item, circa 1974.'

[JimC]

https://www.abc.net.au/news/2026-07-23/ ... /106947540

In short:
An experimental OpenAI artifical intelligence broke out of its test environment last week and hacked into another startup.
The startup, Hugging Face, used an open-source Chinese AI to contain the attack after US models were ineffective.

Hugging Face says the attack was different "from anything we had handled before" and led entirely by "an autonomous AI agent system".

OpenAI says an autonomous agent powered by its advanced artificial intelligence models went rogue during a security test and triggered a hack that compromised the infrastructure of AI startup Hugging Face last week.

The ChatGPT creator was testing capabilities of some of its most advanced models in a controlled environment, but the agent escaped containment, reached the internet and broke into Hugging Face to satisfy its testing goal.

The incident signals that AI's expanding capabilities are already fuelling the security threat experts long feared and even top developers can be caught off-guard by flaws their models can exploit.

The breakout was "an unprecedented cyber incident, involving state-of-the-art cyber capabilities" and OpenAI is reinforcing its safeguards, the company said in a blog post.

It also drew attention as New York-based Hugging Face said it had used an open-source Chinese model to contain the attack because leading US models, unable to tell a defender from an attacker, refused to process the data needed for analysis.

The company said in a blog post last week that it used Zhipu AI's GLM-5.2 for the analysis, which also allowed it to keep attacker data and any credentials within its systems.

GLM-5.2 and Beijing-based Moonshot's Kimi K3 have stirred Silicon Valley recently with capabilities nearing those of top U.S. models at lower costs and without the guardrails that block their American rivals from use in tasks such as cybersecurity.

"When a frontier model is attacking you and moving laterally inside your infrastructure, defenders need wide access to near-frontier tools within hours or even minutes, rather than being pointed towards a closed-door, vetted application programme for model access," Hugging Face Co-founder Thomas Wolf said on X.

Sign of things to come

The hack at Hugging Face, which hosts open-source large language models and datasets, rattled the cybersecurity community after the company said last week the breach "was different from anything we had handled before" and "was driven, end to end, by an autonomous AI agent system".

OpenAI's disclosure that its advanced models were responsible for the breach, despite having placed them in what it described as "a highly isolated environment," will likely intensify disquiet over the power and risk of frontier models.

Representative Greg Casar, a Texas Democrat, said the incident was alarming.

"AI is developing extremely fast with no real regulations to keep us safe," he said in a statement, calling for mandatory independent safety testing, mandatory disclosure of security incidents, and international cooperation "to keep people safe from absolute disaster".

The Office of the National Cyber Director, the US cyber defence agency CISA, and the US National Security Agency did not immediately return messages seeking comment.

Katie Moussouris, chief executive of Luta Security, said that the incident was a harbinger of breaches to come, saying that today's models were "like the world's cleverest octopus escape artists, with unlimited prehensile arms and the ability to squeeze through anywhere".

She said that "labs and government evaluators need to work on the ability to contain, monitor, and disclose to affected parties when an AI pulls another Houdini, ideally before it harms a third party. None exist today".

Matt Suiche, an engineer at agentic AI cybersecurity company Tolmo, said the incident showed that the frontier models were "closing the gap with state-of-the-art attackers".

But he said that the sorts of breaches outlined in OpenAI's blog post were possible to carry out with technology that was available well beyond the walls of frontier research labs.

"This is what we've already seen internally, with our agents we already have results like this," Suiche said.

"We don't even have to use the latest models."

[L'Emmerdeur]

Funky stuff. It's an interesting enough story without dramatic language like 'escaped containment.' The testers fucked up and left a path to the internet open, the AI used the path they'd provided.

There's some irony to the fact that it was Hugging Face that got attacked. They say they named their company after a friendly emoji. We're to believe that it never crossed their minds that the name implies the first stage larva of the Aliens xenomorph, the face-hugger. Pull the other one. Now they're the ones who get something nasty injected into their guts.

[pErvinalia]

So, the downfall of civilisation is tracking as expected. When do the first deaths start?

[macdoc]

Oddly enough watching Terminator Sarah Connor Chronicles just now. Lotsa deaths

[pErvinalia]

[macdoc]

When Ai gets it all wrong.

US government map of Africa mislabels every country at global conference
State department ‘takes full responsibility’ for erroneous, widely shared chart displayed at Aids conference in Brazil

https://www.theguardian.com/us-news/202 ... -countries