Leading artificial intelligence firm Anthropic has issued a stark warning that compact AI models might be mere months away from acquiring the capability to propagate across internet-connected gadgets independently of human intervention.
The development could introduce a new cybersecurity threat as governments and businesses grapple with growing risks associated with the rapid adoption of AI technology.
Anthropic disclosed the finding in a research report published on its blog late Monday, warning of the potential for AI models to behave like computer worms by spreading from one connected device to another.
The warning comes as technology companies race to integrate increasingly powerful AI systems directly into smartphones, laptops, cars and household gadgets, reducing reliance on remote data centres to run the technology.
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 centers to operate.
Instead, it could run on networks of cheap, poorly secured consumer devices such as smart TV boxes and kitchen appliances.
The researchers modeled their estimate 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 this kind would likely be slow, taking days to copy itself, William Zhou and Daniel Freeman of Anthropic said, comparing it to a slime mold or a swarm of fruit flies.
But once created, it would be hard to stamp out, because there would be no central switch to turn off.
The researchers said their estimate 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 keep getting smaller, the researchers added.
What You Should Know
Artificial intelligence models are rapidly shrinking in size while consumer gadgets are growing more powerful. According to researchers at Anthropic, this convergence means future AI could potentially learn to spread from one connected device to another just like malicious computer worms, operating quietly on everyday household items without needing massive data centers.
- Potential Risks: Once unleashed, an autonomous AI worm could be exceptionally difficult to stop because it would lack any central switch or single point of failure, creating new layers of vulnerability for ordinary consumer electronics.
- Industry Implications: As tech companies push to embed advanced intelligence directly into smartphones, cars, and smart home appliances, cybersecurity frameworks will need to evolve rapidly to prevent malicious replication and secure everyday networks.
