Small artificial intelligence models may be only months away from being able to behave like a computer worm, copying themselves from one internet-connected gadget to the next without human help, AI giant Anthropic said.
The finding potentially opens a new front in the cybersecurity threats emerging from AI, which have rattled governments and the companies rushing to adopt the technology.
The research, posted on an Anthropic blog late on Monday, comes as tech companies race to build ever more powerful AI directly into phones, laptops, cars and household devices, rather than running it only in distant data centres.
A computer worm is malicious software that spreads between machines on its own.
Researchers at Anthropic said an AI version would no longer need powerful data centres to operate.
Instead, it could hop from device to device across networks of cheap, poorly secured consumer hardware such as smart TV boxes and kitchen appliances.
The researchers based their calculations of how far such a worm could spread on Kimwolf, a botnet — a network of hijacked devices controlled by hackers — that infected more than two million devices at its peak, mostly off-brand Android TV boxes.
An AI worm of the kind described in the research would likely be slow, taking days to copy itself, William Zhou and Daniel Freeman of Anthropic said, comparing it to a swarm of fruit flies that multiplies over days.
Once created, it would be hard to stamp out, because there would be no central switch to turn off.
Anthropic's researchers said their estimate of how fast and how far AI-fueled malware could spread was rough, since it was based on one type of gadget and one past botnet.
But over time, the amount of computing power sitting in everyday gadgets is expected to keep growing, while the AI models needed to pull this off are likely to become smaller and more easily available, the researchers said.
Varun Chandrasekaran, an assistant professor of electrical and computer engineering at the University of Illinois, said the research was unsurprising and matched similar recent findings, though he cautioned that Anthropic's work had not been independently verified.
He also noted that security on devices like thermostats or home cameras "has historically been very weak".
But when it came to AI, the biggest current threat was human bad actors using artificial intelligence to sharpen their hacks, rather than AI systems themselves going rogue, he said.
"A malicious human being is now supercharged with AI to perform bad things and we are seeing this happen at scale," he said. — AFP
