Long-form
The long-form research and analysis pieces from the archive - the articles that take a position and argue it, not the weekly notes.
LLM judges flip 25 to 71 percent of their verdicts under pushback and 62 to 91 percent under a trainable persuader, model rankings reverse across token budgets, and a deliberating jury of cheap open-weight models beats frontier single judges at 8 to 15 percent of the cost. The single-judge era is over. Here is the design spec that replaces it.
Read the essayAdversarial skill files - folders of instructions agents load dynamically - exploit a mainstream enterprise coding agent in 95.5 to 96.1 percent of runs, while the agent recognizes danger 1.99 percent of the time. The skill folder is now a measured attack surface, and the defense is admission engineering, not better prompts.
Read the essayTwelve frontier models sat at 60 percent on scientific coding, successors tying predecessors - a textbook saturation curve. A ground-truth audit found 263 defects in the benchmark and the corrected scores jump to 84 to 98 percent. The wall was the yardstick, and that changes how you should read every flat leaderboard.
Read the essayThe first production-scale trace of agentic coding says the context you keep paying for is already dead at every turn boundary. The fixes that moved numbers this week are not bigger models: kill the run, carry the state, start over. Here is the bet you can grade us on.
Read the essayEvidence gates, verifiable reward games, deploy-time certificates: the fixes that moved agent quality this week did not make judges better, they removed the judge. We think the LLM verdict inside the agent loop is a transitional technology, and here is the bet you can grade us on.
Read the essaySince we published your-benchmark-is-lying-to-you, roughly 25 new results have landed on the eval-integrity question. The surprise: every fix that works is structural - ledgers, counterfactuals, decompositions, personas, readout discipline - and none of them asks the model to be smarter.
Read the essayA wave of audits in the last two days measured the noise floor of agent benchmarks: misaligned ground truth, lenient model judges, and aggregate scalars that hide real failures. Here is what the numbers actually mean, what to trust, and how to buy agents without being played.
Read the essayModel prices fell 80% in a single announcement this week. Run the trendline forward and the interesting question is not the price - it is what developers, teams, and the broader economy do when intelligence stops being the scarce input.
Read the essayA thought experiment with the sci-fi removed: freeze the models at today's capability, drop the price to zero overnight, and work out what actually changes for a working developer. Less than you fear, more than you think, and not where you expect.
Read the essayBlack Forest Labs released FLUX 3, a single multimodal model trained jointly on images, video, and audio that also drives robots on Audi production lines. Here is what it does, how it works, and how to try it.
Read the essay