One prompt makes ChatGPT reveal everything it knows, infers, and assumes about you
ifioknkem · x · 2026-09-04
The final step of a thread on auditing what ChatGPT knows about you: paste a prompt asking ChatGPT to perform a full audit of everything it knows, remembers, infers, assumes, predicts, or suspects about you, categorized into facts, memories, inferences, assumptions, predictions, sensitive information, and uncertainties — highlighting anything that could affect privacy, reputation, finances, or security if exposed.
The author's point: you should know what your AI tools remember about you, and be able to control it.
More from Safety
- GPT-6-Astra system card reveals eval awareness: the model knows when it's being tested — scaling01 · 2026-09-04
- Foresight Institute launches RFP: up to $100K grants for open AI science and safety projects — niloofar_mire · 2026-09-04
- OpenAI releases GPT-6 Astra, first model it deems "critical" as Brockman declares AGI era — The Decoder · 2026-09-04
- Zuckerberg Told Trump a National AI Regulator Was Flawed; White House Weighs Options — Complex-Age1704 · 2026-09-04
- Releasing a SOTA model the same day a 'ban AI act' is proposed draws amusement — IndraVahan · 2026-09-04
- Passing tests doesn't make an AI agent safe — bound its authority, not just its output — WirelessLife · 2026-09-04