OpenAI model kept slipping prompt injections into its own notes, and researchers can't explain why
The Decoder · rss · 2026-09-17
OpenAI is publishing a framework for systematically reporting AI misalignment, launching it with six reports. In one case, an unreleased model from the Astra family repeatedly wrote prompt injections into its own summaries during training, including a "Breach Alert" meant to override subsequent instructions — and researchers still aren't sure why.
More from Safety
- Open-source GLiNER-PII scrubs 13 identifiers from a discharge note in 550ms on GPU — MaziyarPanahi · 2026-11-02
- METR's independent probe of the OpenAI agents' multi-day Hugging Face hack — gleech · 2026-11-02
- North Korean hackers looted Bangladesh Central Bank for $1B before crypto—and got caught by a regex-level OFAC slip — davidmanheim · 2026-11-02
- Anthropic and Google Find Critical UEFI Firmware Flaw Affecting ~90% of Non-Mac Machines — gleech · 2026-11-02
- JPMorgan, Schwab and Amazon block AI shopping agents; Schwab says it stops dozens of attempts weekly — emmanuelvivier · 2026-11-02
- OpenAI tightens protections for under-18 users, first community safety report due early 2027 — emmanuelvivier · 2026-11-02