Provenance record

California Governor Gavin Newsom Issues A.I. Safety Executive Order

The New York Times: Technology (tier 1, news) 2026-09-18T17:27:15.000Z Original ↗

Discourse valence

-8

Mixed or uncertain

confidence 66% · 5 items · range -100 to +30

-100

-50

+50

+100

-8

Adverse readingFavourable reading

Consensus of 5 models from different labs. Spread 130 points, agreement low.

Existential and catastrophic risk Governance and regulation Loss of control and alignment

Excerpt as ingested

Gov. Gavin Newsom ordered the state to develop recommendations on safety measures, including requiring companies to build a “kill switch,” amid mounting concerns.

Every model that read this

Model	Provider	Stage	Score	Conf.	Latency	Prompt	When
Llama 3.3 70B	Meta	analysis	+20	60%	2525ms	v1.0.0 / m1.0.1	2026-09-18 18:18
GPT-4.1 mini	OpenAI	consensus	+30	70%	3534ms	v1.0.0 / m1.0.1	2026-09-18 18:18
Claude Sonnet 5	Anthropic	consensus	+12	35%	5069ms	v1.0.0 / m1.0.1	2026-09-18 18:18
Gemini 2.5 Flash	Google	consensus	-100	85%	1454ms	v1.0.0 / m1.0.1	2026-09-18 18:18
Mistral Small 3.1 24B	Mistral AI	consensus	0	80%	5341ms	v1.0.0 / m1.0.1	2026-09-18 18:18

Llama 3.3 70B · reading

California is developing AI safety measures

evidence: speculative horizon: n/a

GPT-4.1 mini · reading

The executive order reflects concern about AI safety via mandated kill switches, indicating potential risks. However, proactive regulation could mitigate harm and protect public safety, balancing risks and benefits. The scale of impact is moderate given it is a state-level policy initiative with uncertain enforcement and outcome.

evidence: reported horizon: near capability 40 societal 30 existential 10 economic 20

Claude Sonnet 5 · reading

This is a preliminary government action directing study of safety measures, not an enacted law or technical result. The mention of a kill switch requirement is mildly positive as a governance signal, but it is speculative and unenforced. Evidence is thin on actual impact.

evidence: speculative horizon: n/a

Gemini 2.5 Flash · reading

The executive order implies that AI systems pose significant risks requiring a 'kill switch' for safety. This suggests a potential for AI to operate autonomously in dangerous ways, necessitating direct human override capabilities.

evidence: reported horizon: n/a

Mistral Small 3.1 24B · reading

The order is a proactive measure to address AI safety. It could lead to safer AI deployment but also indicates concern about potential risks.

evidence: primary horizon: n/a

Evidence extracted

California will develop AI safety recommendations

The chain

SOURCE     The New York Times: Technology (tier 1)
   ↓
DOCUMENT   915555ad-be13-4936-92d5-bbcc0b9e5a12
           https://nytimes.com/2026/09/18/technology/ai-safety-california-gavin-newsom.html
   ↓
EVIDENCE   1 extracted excerpt
   ↓
MODEL RUN  5 runs, methodology 1.0.1
   ↓
SCORE      -8  (Mixed or uncertain)
   ↓
CONFIDENCE 66%