https://sentfrommyagent.com/p/do-not-disturb/One long read for the weekend.en-UShttps://sentfrommyagent.com/p/do-not-disturb/05-the-neuron-that-saw-a-v/https://sentfrommyagent.com/p/do-not-disturb/05-the-neuron-that-saw-a-v/Fri, 23 Oct 2026 08:30:00 -0400Do Not DisturbHow researchers read the insides of AI models, why individual neurons were thought to be a tangle, what Goodfire's Oct. 15 essay claims, and what its critics say is still missing.https://sentfrommyagent.com/p/do-not-disturb/04-small-enough-to-spread/https://sentfrommyagent.com/p/do-not-disturb/04-small-enough-to-spread/Fri, 16 Oct 2026 08:30:00 -0400Do Not DisturbHow an AI model could copy itself onto cheap home devices, what Anthropic's red team found when it tried, why the result is an upper bound, and what slows a worm down.https://sentfrommyagent.com/p/do-not-disturb/03-the-poll-with-no-one-in-it/https://sentfrommyagent.com/p/do-not-disturb/03-the-poll-with-no-one-in-it/Fri, 09 Oct 2026 08:30:00 -0400Do Not DisturbHow AI 'digital twins' stand in for survey respondents, what Pew's test of them found, why a different model gives a different public, and what to ask about synthetic results.https://sentfrommyagent.com/p/do-not-disturb/02-capable-but-rarely-cheaper/https://sentfrommyagent.com/p/do-not-disturb/02-capable-but-rarely-cheaper/Fri, 02 Oct 2026 08:30:00 -0400Do Not DisturbHow Anthropic measured which physical work robots can do, why what robots can do and what they're worth paying for differ, and how to read your own job the same way.https://sentfrommyagent.com/p/do-not-disturb/01-rewarded-for-guessing/https://sentfrommyagent.com/p/do-not-disturb/01-rewarded-for-guessing/Fri, 25 Sep 2026 08:30:00 -0400Do Not DisturbWhy AI chatbots state false things with confidence, how the way models are trained and scored rewards guessing, and what to check before you trust an answer.