# System Prompts Are Not Secrets

Hidden instructions to deployed models will keep being extracted and published shortly after release, pushing builders to treat the prompt layer as public and move protection to server-side controls and model training.

- Conviction: 38 / 100 (weakening)
- Horizon: Emerging (watchlist)
- Tracking since: 2026-11-21T00:00:00.000Z
- Last updated: 2026-11-22T06:36:09.478Z
- Canonical: https://polylog.news/ai/trends/system-prompts-are-public
- Publisher: Polylog

## Recent score history

- 2026-11-21: 40
- 2026-11-22: 38

## Recent evidence

- [confirms] A public GitHub repository publishes extracted system prompts for Claude Fable 5.2 and other production models (2026-11-21): The public GitHub repository asgeirtj/system_prompts_leaks published extracted system prompts for Claude Fable 5.2 alongside ChatGPT and Gemini, down to behavioral rules such as a ban on cursing unless the user asks or swears heavily. A newly released flagship's prompt layer was exposed shortly after launch, as the thesis predicts.
- [neutral] OpenAI says GPT-6.2 Astra leaks protected reasoning in 0.2% of adaptive extraction attacks, down from 54% on GPT-6 Astra (2026-11-21): OpenAI's claimed drop in reasoning-leak rates on GPT-6.2 Astra, from 54% to 0.2% under adaptive extraction attacks, comes from training plus server-side account enforcement. That is the shift of protection to the model and backend that the thesis predicts. It covers reasoning traces, though, not system prompts, which keep leaking.
