yupanqui

Petrov Day in Cusco, with a candle

AI with per­sis­tent con­text caught a self-pun­ish­ment pat­tern I’d hid­den from myself

Ca­pa­bil­ity didn’t fix my avoidance; it made the avoidance more sophisticated

• yupanqui 21 Jan 2026 15:27 UTC

  1 point

  0

  in reply to: Aprillion’s comment on: So Long Sucker: AI De­cep­tion, “Alli­ance Banks,” and In­sti­tu­tional Lying

  I haven’t found research on game length and betrayal timing in humans specifically. The closest is iterated prisoner’s dilemma work on end-game effects. If you find anything, I’d be curious, would help clarify if this is LLM-specific or mirrors human behavior

• yupanqui 21 Jan 2026 15:25 UTC

  1 point

  0

  in reply to: lilkim2025’s comment on: So Long Sucker: AI De­cep­tion, “Alli­ance Banks,” and In­sti­tu­tional Lying

  You’re right, Kimi struggled with state tracking. That’s partly why the research focused on Gemini vs GPT-OSS comparisons.
  On running a broader tournament: I’d love to, but API costs are the constraint. The current version now includes more Gemini models (just added). For a proper ELO-style tournament across frontier models, I’d need to figure out funding or find collaborators with API credits. If anyone wants to run games with other models, the code is open source and logs everything.

So Long Sucker: AI De­cep­tion, “Alli­ance Banks,” and In­sti­tu­tional Lying

Song: Party In The AI Lab

• yupanqui 10 Oct 2025 23:58 UTC

  2 points

  1

  on: The Story of “I Have Been A Good Bing”

  Here is a new one for you guys:
  

• yupanqui 4 Aug 2025 20:15 UTC

  3 points

  2

  on: Should we aim for flour­ish­ing over mere sur­vival? The Bet­ter Fu­tures se­ries.

  i like the name viatopia , but perhaps search for a clear and simple name, like opentopia.

How I Spent 2024 Liv­ing Like the World Was Go­ing to End