Provenance record

OpenAI sued by Canadian province over role in mass shooting

Financial Times: Artificial Intelligence (tier 1, news) 2026-09-22T01:28:52.000Z Original ↗

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

Discourse valence

-70

Strongly adverse

confidence 55% · 5 items · range -100 to -45

-100

-50

+50

+100

-70

Adverse readingFavourable reading

Consensus of 5 models from different labs. Spread 55 points, agreement low.

Existential and catastrophic risk Loss of control and alignment Concentration of power

Excerpt as ingested

British Columbia claims the ChatGPT maker failed to act after attacker’s account was flagged

Every model that read this

Model	Provider	Stage	Score	Conf.	Latency	Prompt	When
Llama 3.3 70B	Meta	analysis	-75	80%	2775ms	v1.0.0 / m1.0.1	2026-09-22 03:17
GPT-4.1 mini	OpenAI	consensus	-65	80%	3974ms	v1.0.0 / m1.0.1	2026-09-22 03:17
Claude Sonnet 5	Anthropic	consensus	-45	35%	7961ms	v1.0.0 / m1.0.1	2026-09-22 03:17
Gemini 2.5 Flash	Google	consensus	-100	0%	1958ms	v1.0.0 / m1.0.1	2026-09-22 03:18
Mistral Small 3.1 24B	Mistral AI	consensus	-65	80%	8104ms	v1.0.0 / m1.0.1	2026-09-22 03:18

Llama 3.3 70B · reading

AI linked to mass shooting

evidence: reported horizon: n/a

GPT-4.1 mini · reading

The lawsuit alleges failure of an AI company to prevent harm despite warnings, highlighting risks of AI misuse in violent acts and accountability challenges. This implies significant negative societal impact but lacks detailed outcomes or systemic evidence.

evidence: reported horizon: near capability 50 societal 70 existential 10 economic 40

Claude Sonnet 5 · reading

This is an unproven legal allegation reported only via headline and standfirst, with no detail on what was flagged or how OpenAI allegedly failed to respond. The implication is materially concerning if true, involving a failure to prevent real-world violence, but evidence is thin and unresolved.

evidence: speculative horizon: n/a

Gemini 2.5 Flash · reading

evidence: reported horizon: n/a

Mistral Small 3.1 24B · reading

A lawsuit alleging negligence in preventing harm raises serious concerns. It suggests AI systems may enable harm if not properly managed.

evidence: primary horizon: n/a

Evidence extracted

OpenAI sued by Canadian province
OpenAI failed to act after attacker's account was flagged

The chain

SOURCE     Financial Times: Artificial Intelligence (tier 1)
   ↓
DOCUMENT   812cd673-4647-4da8-a046-0a36e0148fdf
           https://ft.com/content/30d2f29b-e750-4a90-9e8d-b1a7f14e4908
   ↓
EVIDENCE   2 extracted excerpts
   ↓
MODEL RUN  5 runs, methodology 1.0.1
   ↓
SCORE      -70  (Strongly adverse)
   ↓
CONFIDENCE 55%