ChatGPT Reasoning Trace Extraction Vulnerability Patched
wunderwuzzi23 · x · 2026-08-12
Security researcher @wunderwuzzi23 provided an update on the recently exposed ChatGPT reasoning trace extraction vulnerability:
- Previous Exploit: A few hours ago, he managed to recover reasoning traces from another account using a specific endpoint that had not been patched yet.
- Fully Patched: OpenAI has now fixed the issue. Extracting traces—even from his own account with a new session—no longer works.
- Next Steps: He has recorded videos of the exploit and will continue monitoring the situation.
More from Safety
- Bypassing Claude's Invisible Watermark: Free Rewriting Tool Launches — MatthewChang · 2026-08-12
- The Guardrail Tax: Enterprise AI Safety Overhead Costs More Compute Than Reasoning — vasilisvj · 2026-08-12
- Unspecified SSH Username Prompts Claude Agent to Brute-Force and Get Banned — SebastianNehrd2 · 2026-08-12
- Chinese Farmer Loses 25 Acres of Sesame After AI Recommends Fatal Chemical Mix — Polymarket · 2026-08-12
- Proving Personhood Online: The Challenge of AI Agents Roaming the Web — SuB8u · 2026-08-12
- Vulnerability in Major LLM APIs Exposes Encrypted Reasoning and Leaks Passwords — yangyi · 2026-08-12