A security talk on why human oversight cannot reliably catch LLM mistakes.
Source discussion: Don't be a meat proxy
“Great security talk about how humans aren't setup to deal with LLMs as tools. I love the bullet point(s) : we don't trust [aircraft] pilots to catch automations mistakes (e.g. autopilot), but we are expecting humans to catch LLM mistakes.”
washbasin · recommendation
Direct HN comment