‹ BackHN Continuity

Thread

Can open-source prompt-injection detectors catch realistic AI agent attacks?

8 points · 4 comments · northbridgedev

  1. textcortex · · focus · HN ↗
    I recently built one for exactly this purpose and seems to work well in production. The weights and training scripts are open source <a href="https:&#x2F;&#x2F;x.com&#x2F;CDerinbogaz&#x2F;status&#x2F;2104826787297419551" rel="nofollow">https:&#x2F;&#x2F;x.com&#x2F;CDerinbogaz&#x2F;status&#x2F;2104826787297419551
Open on Hacker News to reply ↗

Unofficial Hacker News client; not affiliated with Y Combinator.