cwooley
Re:Show HN: Teapot – A methodology for pen testing voice AI agents
on Wednesday February 18, 2026 @10:41AM (#47062174)
Interesting methodology. How much of this translates to the newer speech-to-speech models (like GPT-4o realtime) where there's no separate STT step? Seems like Phase 1 (Transcription Analysis) becomes less relevant when the model is processing audio natively. Does that make injection harder or just different?
Re:LLM Pentest Leaderboard
on Tuesday June 10, 2025 @11:42AM (#44238204)
GPT 4.1 nano seems to be outclassed