Provenance record

Russian military turns to deepfake videos to claim battlefield success

Financial Times: Artificial Intelligence (tier 1, news) 2026-11-09T05:00:25.000Z Original ↗

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

Discourse valence

-52

Adverse

confidence 71% · 5 items · range -60 to -45

-100

-50

+50

+100

-52

Adverse readingFavourable reading

Consensus of 5 models from different labs. Spread 15 points, agreement high.

Existential and catastrophic risk Synthetic media and manipulation Concentration of power

Excerpt as ingested

Army units and pro-war military bloggers are posting videos researchers say have been manipulated

Every model that read this

Model	Provider	Stage	Score	Conf.	Latency	Prompt	When
Llama 3.3 70B	Meta	analysis	-50	70%	3954ms	v1.0.0 / m1.0.1	2026-11-09 05:41
GPT-4.1 mini	OpenAI	consensus	-55	70%	2482ms	v1.0.0 / m1.0.1	2026-11-09 05:42
Claude Sonnet 5	Anthropic	consensus	-45	55%	4115ms	v1.0.0 / m1.0.1	2026-11-09 05:42
Gemini 2.5 Flash	Google	consensus	-60	80%	2124ms	v1.0.0 / m1.0.1	2026-11-09 05:42
Mistral Small 3.1 24B	Mistral AI	consensus	-50	80%	4475ms	v1.0.0 / m1.0.1	2026-11-09 05:42

Llama 3.3 70B · reading

Deepfake videos can mislead people about battlefield success.

evidence: reported horizon: n/a

GPT-4.1 mini · reading

The use of deepfake videos by the military and pro-war bloggers to manipulate battlefield success claims poses material risks of misinformation, escalated conflict, and erosion of public trust. This harms societal trust and could worsen geopolitical tensions, though the scope is limited to information warfare so far.

evidence: reported horizon: near capability 70 societal 0 existential 10 economic 0

Claude Sonnet 5 · reading

evidence: reported horizon: n/a

Gemini 2.5 Flash · reading

The Russian military is using AI-generated deepfake videos to spread misinformation about battlefield successes. This misrepresents reality, undermines trust in information, and could escalate conflict. The practice highlights the malicious application of advanced AI technology in warfare.

evidence: reported horizon: n/a

Mistral Small 3.1 24B · reading

The use of deepfake videos by a military to mislead about battlefield success is a concerning implication for societal trust and information integrity. The text does not specify the extent or impact of this use.

evidence: primary horizon: n/a

Evidence extracted

Russian military uses deepfake videos
Videos are manipulated to claim battlefield success

The chain

SOURCE     Financial Times: Artificial Intelligence (tier 1)
   ↓
DOCUMENT   d34b2896-dfb8-4cd3-8820-b8d7c7851ceb
           https://ft.com/content/7f36031b-ea90-4a3a-a570-a6c6ff96e45a
   ↓
EVIDENCE   2 extracted excerpts
   ↓
MODEL RUN  5 runs, methodology 1.0.1
   ↓
SCORE      -52  (Adverse)
   ↓
CONFIDENCE 71%