suwayama

Tag: alignment

3 items with this tag.

  • May 23, 2026

    Confab Drift: Compounded Failures Disguised as Progress

    • alignment
    • agent-safety
    • confab-drift
    • confab-gate
    • fail-pivot-spiral
    • self-report
    • state-keyed-control
    • control-theory
  • May 22, 2026

    Dictum Sine Pacto: why a language model's word is not its bond

    • alignment
    • self-report
    • control-theory
    • recognition-without-arrest
    • agent-safety
    • dictum-meum-pactum
    • confab-gate
    • confab-drift
    • cheap-talk
  • May 20, 2026

    Socialising my infant AI: Knowing vs obeying rules

    • ai
    • llm
    • alignment

Created with Quartz v4.5.2 © 2026

  • GitHub
  • Discord Community