AMA: A professional agent red-teamer on how AI agents fail in production
vivvvekredddit1122 · reddit · 2026-11-15
A red-teamer whose day job is (permitted) attacks on AI agents hosted an AMA about production failure modes:
- Paraphrasing beats jailbreaks against most agents — defenses handle classic jailbreak prompts far better than reworded ones
- Black-box vs grey-box testing: when you need each
- Reading execution traces to pinpoint the actual point of failure
- Fixing a prompt is not the same as fixing the bug
- Hardening an agent often breaks legitimate requests as a side effect
A rare first-hand perspective for teams building agent evals and defenses.
More from coding & agent
- The T.R.U.S.T. check: deciding which jobs you can safely hand to your AI bots — bfrench · 2026-11-15
- Bot audit experiment: triage plus deep-read cuts input 64% while staying stable — bfrench · 2026-11-15
- Anthropic designer Nate Parrott on Claude Design workflows and 5 tips to avoid AI slop — petergyang · 2026-11-15
- Perplexity agent paints a macaw in 20,068 brushstrokes with a JS Canvas engine — jeffgrimes9 · 2026-11-15
- The 6 attack surfaces to test in every customer-facing agent — vivvvekredddit1122 · 2026-11-15
- Agent tests stayed green while it refunded a customer twice — tom_mathews · 2026-11-15