Fetching from the wire…
01
02
03
04
05
06
07
08
Security2026-06-02 · source-backed
Prompt injection and jailbreaks often have no clean patch, and automated tooling rediscovers them in hours. He proposes a tiered model: immediate vendor notification, 30-day public disclosure for prompt-level flaws, indefinite embargo for model-weight attacks where no patch exists (Schneier). He's right that the old model assumes a patchable defect. A lot of AI flaws aren't defects, they're properties.
Each link below shares sources, entities, or timing with this story.
Shared entity: Prompt / Shared topic / What happened next / Tension
Both cover Prompt; overlapping topics (attack, model); picks up the Prompt thread on 2026-07-28.
Shared entity: Prompt / Shared topic / What happened next
Both cover Prompt; overlapping topics (attack, have, model); picks up the Prompt thread on 2026-06-27.
Both cover Prompt; overlapping topics (hour, model); picks up the Prompt thread on 2026-08-14.
Shared entity: Schneier / Same source domain / What happened next
Both cover Schneier; reported by the same outlet (schneier.com); picks up the Schneier thread on 2026-07-31.
Both cover Schneier; reported by the same outlet (schneier.com); picks up the Schneier thread on 2026-07-26.
Shared entity: Prompt / Shared topic / What happened next
Both cover Prompt; overlapping topics (have, model); picks up the Prompt thread on 2026-06-30.
Shared entity: Schneier / Same source domain / Earlier coverage
Both cover Schneier; reported by the same outlet (schneier.com); earlier Schneier coverage from 2026-03-10.
Both cover Schneier; reported by the same outlet (schneier.com); earlier Schneier coverage from 2026-03-10.