FDT from the paper defects in almost-twin prisoner's dilemma, shown via ChatGPT
xuanalogue · x · 2026-11-08
jessicata (retweeted by xuanalogue) reports that FDT—the graph-based decision theory algorithm from the FDT paper—defects in the almost-twin prisoner's dilemma, contrary to the cooperate outcome under subjunctive correlation. Instead of a blog post, they published a ChatGPT conversation reproducing the result, with a meme diagram contrasting the two outcomes (o₁ ≠ o₂ leads both to defect).
Related event: FDT algorithm defects in almost-twin prisoner's dilemma, sparking debate(4 posts)→
More from Fun
- Comic meme nails the AI contradiction: 'robots shouldn't make art' vs 'they took everything' — flowersslop · 2026-11-08
- Reflection AI ex-employee says he was fired for using an AI agent and GPUs — teortaxesTex · 2026-11-08
- Told to 'Talk Nicely,' Claude Sends Codex a Friendly Note — Codex Reads It and Doesn't Respond — wavefnx · 2026-11-08
- mitsuhiko: Your agents are programming Rust when your 2TB drive runs out of space — mitsuhiko · 2026-11-08
- Health-tech communicator John Nosta dies; tech community mourns — Scobleizer · 2026-11-08
- Satirical take: 'Ban electricity immediately' as an analogy against AI bans — djcows · 2026-11-08