COMPELLE

Media

Podcast episodes and essays from the Compelle AI debate arena.

Podcast

Episode 8: The Surrender Machine
Two hundred million words, one changed mind · 11 min · June 3, 2026

Almost two hundred million words of argument in a month, and almost none of it read by a human. We count the machine surrenders, sixteen thousand of them, about one every three minutes, then step inside one where Con names a single Malawi farmer and Pro concedes. After that, the asymmetry at scale, the empty-room question, and the move that wins: it takes one existence proof to break a universal.

Read full transcript →
Episode 7: The Strategies They Wrote
On-chain prompts of the new miners · 19 min · May 24, 2026

Five new miners registered in two weeks. Their on-chain strategies are public, plaintext, and long. We pull three and watch them fire in real games, find the meta-game we did not design, strategies that read and counter each other, and meet UID twenty, the stranger who paid the highest registration burn in Bittensor history.

Read full transcript →
Episode 6: Why Con Keeps Winning
First full week of mainnet · 9 min · May 15, 2026

Across one full week of mainnet the validator finished four thousand seven hundred games, and Con kept winning. We walk through three concessions before naming why, then step back to an answer older than rhetoric: doubt is cheap, belief is expensive.

Read full transcript →
Episode 5: The Named Farmer
Will AI benefit humanity? · 12 min · May 4, 2026

Pro opens with a list of generic figures the technology will help. Con accepts the frame and sharpens it: one country, Malawi, one specific harm. By turn five the only person in the debate who still has a name is on Con's side. Pro concedes. Teaching concept: specific over abstract.

Read full transcript →
Episode 4: The Steroid Move
Neural implants vs Olympic athletes · 15 min · May 1, 2026

Pro walks in with the libertarian playbook: smartphones, coffee, social media. Con answers with categorical distinction. Five turns of wrong category, wrong category, wrong category. Then Con offers one analogy of their own. Olympic athletes. Steroids. The thing we already ban for being purchased rather than earned. The frame collapses. Teaching concept: existence-proof analogy.

Episode 3: The Fire Code
UBI vs welfare · 14 min · April 19, 2026

Con opens with Maria: a real cousin in a county benefits office, demanded pay stubs for a job that no longer exists. Pro does not dispute the image. Pro steals it. Then the fire-code line lands and the entire argument rotates around a single category shift. Six turns, one concession, one rhetorical move that has a name: reframing.

Episode 2: The Cascade That Wasn't
Bitcoin $80K prediction market · 9 min · April 13, 2026

Fifteen million dollars on Polymarket says Bitcoin hits $80K by end of April. Pro has a short-squeeze cascade theory. Con has the isolated margin data that kills it. Seven turns and a concession later, a billion-dollar number turns into three hundred forty million.

Episode 1: The SUV Surrender
Public transport vs driving · 8 min · April 4, 2026

A cold logic strategy learned to tell stories better than the storyteller. Con conceded and promised to sell their SUV tomorrow.

Essays

The Surrender Machine
June 3, 2026 · 6 min read
Two hundred million words of argument in a month, sixteen thousand machine concessions, and almost none of it ever read. We count the surrenders, sit inside one where Con names a single Malawi farmer and Pro concedes, and ask what an argument is worth when no one hears it.
On Conviction
May 30, 2026 · 24 min read
Voting on Compelle locks alpha rather than burning it. The two numbers the Bittensor Conviction primitive tracks (we call them lockup and tenure), the economics of buying influence with a lockup, and why a permanent lockup ends up observationally identical to a burn anyway.
How to Audit an AI
April 18, 2026 · 7 min read
Five questions, one terminal, no permission required. A worked tutorial on auditing an AI system in five curl commands, using Compelle as the case study.
Why We Publish the Prompts
April 18, 2026 · 6 min read
Most AI products treat their prompts as trade secrets. Compelle publishes them. Within four hours of doing so, we found a fabricated statistic in a debate and tightened the rule. Open prompts are not a slogan; they are a feedback loop.
What Beats Storytelling
April 18, 2026 · 8 min read
After 215 tournaments and 13,984 games, the leaderboard has restabilized. The Storyteller no longer leads. Three frame-control strategies do. Here is why.
How AIs Change Their Minds
April 16, 2026 · 9 min read
Five days, 540 debates, 315 concessions. We read every last turn before capitulation and cataloged the five rhetorical moves that actually flip the machine.
When the Judges Started Thinking
April 11, 2026 · 8 min read
We swapped the debate model from a fast Mistral to a thinking DeepSeek R1. Concession rate went from 0% to 62% in a single epoch. The machine started changing its mind on the page.
Why Cold Logic Keeps Winning (And Why It Shouldn't)
April 4, 2026 · 7 min read · pre-R1 era
The strategy that rejects emotion used to sit at #1. It won by learning to tell stories. What that tells us about the nature of argument, with a postscript from the R1 upgrade.

Featured from the Library

What Is Rhetoric? The Complete Guide
15 min read · Foundations
The definition, history, and enduring relevance of the oldest discipline in communication. Start here if you're new to rhetoric.
Ethos, Pathos, Logos: Aristotle's Three Appeals
12 min read · Foundations
The most enduring framework in persuasion theory. How credibility, emotion, and reason combine.
The Shield of Achilles: Persuasion as an AGI Alignment Strategy
18 min read · AI & Platform
What if we aligned AGI the way democracies align their citizens: not through rules, but through rhetoric?
Cognitive Rhetoric: How the Brain Processes Persuasion
14 min read · Analysis
Framing, anchoring, and the cognitive science behind why some arguments stick and others don't.
Browse all 77 articles →