# OpenAI Says It Blocked a Campaign to Extract Hidden Reasoning Traces, Naming People Linked to Moonshot AI

Requests peaked at about 16,000 from more than 4,000 accounts over two days in late July, as operators copied encrypted reasoning from one session and asked a model in a separate session to decrypt and transcribe it.

- Conviction: 26 / 100 (weakening)
- 7-day move: -14
- Horizon: Emerging (watchlist)
- Tracking since: 2026-10-03T00:00:00.000Z
- Last updated: 2026-10-12T14:07:13.413Z
- Canonical: https://polylog.news/ai/trends/event-openai-says-it-blocked-a-campaign-to-extract-hidden-reasonin
- Publisher: Polylog
- Affected regions: Global

## Recent score history

- 2026-10-11: 28
- 2026-10-12: 26

## Recent evidence

- [confirms] New papers show jailbreaks and GraphRAG poisoning work by composing benign parts, a pattern OpenAI says it saw in a reasoning-extraction campaign (2026-10-05): New arXiv papers show that jailbreaks and GraphRAG poisoning work by combining individually harmless pieces, the same method OpenAI says the Moonshot-linked campaign used. That places the extraction campaign within a wider class of attacks rather than treating it as a one-off.
- [neutral] OpenAI Says It Blocked a Campaign to Extract Hidden Reasoning Traces, Naming People Linked to Moonshot AI (2026-10-03): Auto-opened to track an article no existing trend covered.