What are self-replicating ‘AI worms’, and why is Anthropic worried about them?

The tech company says successful AI worms need more capability on less powerful devices, but current trends make that feasible by the end of the year Warnings about what artificial intelligence might do in the future – together with alarming reports of what it has gotten up to already – are being issued so quickly that even security experts have difficulty triaging what to worry about. Yet a report this week from Anthropic’s offensive research team grabbed attention by describing a new, practical threat, one devoid of human characteristics or apocalyptic musings. The red team, as such internal testers are called, said it created a malicious AI program that spread autonomously through internet-connected “smart” televisions and could do the same inside other ordinary devices. While machines running AI usually need a fair amount of computing power, the researchers said that their program split parts of itself among 15 televisions and was still able to function by using a bit of power from each. Continue reading...

AI worms can spread autonomously through devices, potentially causing harm.

The report describes a practical and novel risk from AI: self-replicating AI worms that can spread autonomously on everyday devices, raising cybersecurity risks and potential harms. Although not catastrophic, this threat can materially undermine security and trust in AI-enabled infrastructure. The evidence is a specific report from a known company and is credible enough to warrant significant concern.

This describes a concrete demonstrated capability (distributed AI malware across 15 TVs) rather than pure speculation, which is more concerning than typical warnings. However it is a controlled red-team exercise, not an actual attack, and feasibility timelines are projections.

The article describes Anthropic's offensive research team creating a malicious, self-replicating AI program. This program autonomously spread across internet-connected devices, demonstrating a new practical threat. While the current example is contained, the implication is a severe risk of widespread autonomous cyberattacks, affecting device security and potentially critical infrastructure.

The creation of an AI worm that can spread autonomously and function on less powerful devices is a concerning development. This implies a new vector for malicious AI activity, posing significant risks to cybersecurity and privacy.

SOURCE     The Guardian: Artificial Intelligence (tier 1)
   ↓
DOCUMENT   049c303b-0de9-4015-8859-7f4ad8b49dc6
           https://theguardian.com/technology/2026/oct/18/anthropic-ai-worm-worries
   ↓
EVIDENCE   2 extracted excerpts
   ↓
MODEL RUN  5 runs, methodology 1.0.1
   ↓
SCORE      -66  (Strongly adverse)
   ↓
CONFIDENCE 78%