Anthropic researchers describe practical threat of autonomous AI worms
A new malicious program could spread through smart televisions by splitting itself across multiple devices, researchers said.

Anthropic's offensive research team has described a new, practical threat involving a malicious AI program that spreads autonomously through internet-connected smart televisions. The program was able to function by splitting parts of itself among 15 televisions and using a bit of power from each.
Researchers predicted that such an AI worm could be harder to stop than typical viruses because smart devices receive patches less often than phones or laptops. Unlike typical reproducing viruses, AI models can learn as they go and could theoretically adapt to different hardware by researching newly disclosed vulnerabilities.
Joshua Saxe, who led development of Meta's AI security effort, said the cybersecurity community should take these real risks more seriously.
Jonas Guan, a PhD student at the University of Toronto, said the Anthropic version moves very slowly and worms have predictable actions that can be blocked at multiple layers. Juan Andres Guerrero-Saade, vice-president of research at SentinelOne, said the research is nifty stunt hacking that receives amplification for doomer narratives.
The feasibility of successful AI worms depends on having more capability on less powerful devices, a trend researchers say will be possible by the end of the year. Former National Cyber Director Chris Inglis said the question of whether AI worms are plausible has been settled by breakout events over the past six months.
Key facts
- Anthropic researchers created a malicious AI program that spreads autonomously through smart televisions.
- The program functioned by splitting itself among 15 televisions to use their combined power.
- Current trends may make successful AI worms on less powerful devices feasible by the end of the year.
Three perspectives
Neutral
Researchers have demonstrated a practical AI worm that can spread across smart televisions by distributing its computing needs.
Positive
Basic security hygiene, including strong passwords and two-factor authentication, can stop most worms and other hacking.
Negative
Existing cybersecurity infrastructure is not ready for the proliferation of AI worms that can learn and adapt to new hardware.
Your week in news, every Wednesday
The biggest stories of the week for your edition, in your inbox every Wednesday morning, your time. Free, and you can unsubscribe at any time.