‹ BackHN Continuity

Thread

Why I'm still bearish on LLMs after Navier-Stokes

496 points · 653 comments · jaykru

  1. robinpie · · focus · HN ↗
    I really appreciate seeing a tempered take that's not literally denialist about current capabilities.
    1. brindleth · · focus · HN ↗
      > current frontier models need laborious oversight and guardrails on even the simplest tasks

      It is literally denialist about current capabilities

      1. jaykru · · focus · HN ↗
        why don't anthropic and openai ship yolo mode by default?
        1. Human-Cabbage · · focus · HN ↗
          They do…? Well, “auto” mode has been default in Claude Code for a couple months now. It’s effectively “safer yolo:” tool calls are inspected by a separate classification system (another smaller LLM, I believe) to approve or deny. And you can always layer on additional sandboxing mechanisms to limit the blast radius deterministically.
          1. vmg12 · · focus · HN ↗
            > They do…? Well, “auto” mode has been default in Claude Code for a couple months now

            They have never shipped "yolo" mode by default. Auto mode is not yolo mode. They trained a task specific model just for ensuring the llm didn't accidentally delete every file from your computer.

Open on Hacker News to reply ↗

Unofficial Hacker News client; not affiliated with Y Combinator.