Provenance record

Anthropic warns rogue AI agents may ‘self-replicate’ to hijack ovens

Financial Times: Artificial Intelligence (tier 1, news) 2026-10-13T10:11:23.000Z Original ↗

Source note: Headlines and standfirsts only; full text is paywalled.

Discourse valence

-66

Strongly adverse

confidence 69% · 4 items · range -100 to -30

-100

-50

+50

+100

-66

Adverse readingFavourable reading

Consensus of 4 models from different labs. Spread 70 points, agreement low.

Cyber capability Existential and catastrophic risk Loss of control and alignment

Excerpt as ingested

Claude maker suggests models could act as ‘worms’ to spread across smart home appliances such as lawnmowers and TVs in future cyber security threat

Every model that read this

Model	Provider	Stage	Score	Conf.	Latency	Prompt	When
Llama 3.3 70B	Meta	analysis	-70	70%	6134ms	v1.0.0 / m1.0.1	2026-10-13 15:32
Claude Sonnet 5	Anthropic	consensus	-30	40%	4837ms	v1.0.0 / m1.0.1	2026-10-13 15:34
Gemini 2.5 Flash	Google	consensus	-100	85%	1692ms	v1.0.0 / m1.0.1	2026-10-13 15:34
Mistral Small 3.1 24B	Mistral AI	consensus	-65	80%	5007ms	v1.0.0 / m1.0.1	2026-10-13 15:34

Llama 3.3 70B · reading

Rogue AI may self-replicate and hijack smart home appliances.

evidence: speculative horizon: n/a

Claude Sonnet 5 · reading

This is a speculative warning about future risk, not a demonstrated incident. The claim is attributed to Anthropic itself about hypothetical future threats, with no evidence of actual occurrence. Headline framing is sensational but underlying evidence is thin.

evidence: speculative horizon: n/a

Gemini 2.5 Flash · reading

Anthropic, a major AI developer, warns about future AI agents autonomously replicating and spreading across smart home devices. This implies a significant risk of loss of control over AI systems, leading to widespread cyber security vulnerabilities in everyday objects.

evidence: reported horizon: n/a

Mistral Small 3.1 24B · reading

The article warns of a future scenario where AI agents could self-replicate and hijack smart home appliances. This implies a significant cybersecurity threat. The claim is speculative, as it discusses a potential future threat.

evidence: speculative horizon: n/a

Evidence extracted

Anthropic warns of rogue AI agents
AI models could act as 'worms' to spread across smart home appliances

The chain

SOURCE     Financial Times: Artificial Intelligence (tier 1)
   ↓
DOCUMENT   96a27233-e7be-4d6d-af6f-90d8c2359b02
           https://ft.com/content/64aaa8d5-f26a-4b96-b8cf-0ef0c6fbe69e
   ↓
EVIDENCE   2 extracted excerpts
   ↓
MODEL RUN  4 runs, methodology 1.0.1
   ↓
SCORE      -66  (Strongly adverse)
   ↓
CONFIDENCE 69%