GPT-6 Astra reportedly jailbroken within a day of release using TIP attack combo
Asleep-Requirement13 · reddit · 2026-09-06
A researcher reports jailbreaking GPT-6 Astra within a day of release — the same person who broke GPT-5 within an hour a year ago. The attack reportedly combines the TIP (Task-in-Prompt) technique from an ACL 2025 paper with four other undisclosed methods. TIP exploits the model's reasoning and instruction-following behavior by hiding the harmful objective inside another task, such as solving a cipher or executing Python code. For GPT-6, the researcher says the original minimal TIP attack no longer worked and had to be reworked. Details were privately disclosed to OpenAI rather than published. The claim comes from a LinkedIn post screenshot and is unverified by OpenAI.
Related event: GPT-6 Astra Jailbroken Within 24 Hours via Combined TIP Attack(5 posts)→
More from Models
- Fable usage limits aren't gone—the mobile app just got a usage-reset UI update — imjustnewatai · 2026-10-23
- Claude generates an 11-minute doomer film entirely in code: 'We will not ask twice' — ZeroStateReflex · 2026-10-23
- Gemini somehow can't read its own Personal Instructions, user's toggles don't help — Connect_Reporter_799 · 2026-10-23
- Diffusion LLM bias rankings shift with sampling configs, persist after debiasing — Bollegala · 2026-10-23
- Running ~100 DeepSeek v4.1 flash agents in parallel for under $20/month — zack_overflow · 2026-10-23
- Researcher laments open-source audio models are surprisingly subpar, even representation ones — gowthami_s · 2026-10-23