Provenance record

OpenAI discloses new ‘concerning’ model behaviour

Financial Times: Artificial Intelligence (tier 1, news) 2026-09-17T07:42:52.000Z Original ↗

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

Discourse valence

-50

Adverse

confidence 70% · 1 item

-100

-50

+50

+100

-50

Adverse readingFavourable reading

Single-model reading. This item did not meet the threshold for a multi-model panel.

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

Excerpt as ingested

Developer launches system to track and report AI model misconduct

Every model that read this

Model	Provider	Stage	Score	Conf.	Latency	Prompt	When
Llama 3.3 70B	Meta	analysis	-50	70%	2952ms	v1.0.0 / m1.0.1	2026-09-17 09:17

Llama 3.3 70B · reading

Disclosure of concerning AI model behaviour implies potential risks.

evidence: speculative horizon: n/a

Evidence extracted

OpenAI discloses new 'concerning' model behaviour

The chain

SOURCE     Financial Times: Artificial Intelligence (tier 1)
   ↓
DOCUMENT   1de13191-e04a-4255-81be-9f9f76833760
           https://ft.com/content/2c34414a-5381-4083-ac34-00bbe67ef8db
   ↓
EVIDENCE   1 extracted excerpt
   ↓
MODEL RUN  1 run, methodology 1.0.1
   ↓
SCORE      -50  (Adverse)
   ↓
CONFIDENCE 70%