Hey, I’m Ram. AI and chips, explained from the factory floor.
This weekMES & Shop-Floor Systems3:29
The AI Companies Want an Examiner. Chip Factories Have Had One Since 1973.
AI Just Asked to Slow Down. Chip Makers Wrote the Rule in 1973. The mechanism underneath it, the number that makes it real, and where I could be wrong.

This week’s story
Everything, 45 piecesAI Just Asked to Slow Down. Chip Makers Wrote the Rule in 1973.
The AI companies just asked permission to slow down. On 12 September 2026 Dario Amodei, who runs Anthropic, the company that makes Claude, published an essay asking the whole industry to move one to two years slower so the people who check these systems can catch up, and he offered outside checkers the same access…
Quiz of the week
All weeksThe AI companies just asked for one to two years to let outside checkers catch up. How long did the chip makers give themselves for their next change?
Ten questions from this week in AI and the factory floor. Two are match-ups. Every answer is a fact from a piece on this site, and every explanation links to it.
10 questions, 2 match-ups, about 4 minutes. Every answer links to the piece it came from.
Take the quizPick a door
The guided tour
13 piecesAI on the factory floor
What AI actually changes in a plant, and what it never will.
- AI Just Asked to Slow Down. Chip Makers Wrote the Rule in 1973.
- Anthropic Just Made AI Talk to Machines. Talking Was the Easy Part.
10 pieces · 31 explainersMES & shop-floor systems
The software that actually runs a factory.
- OpenAI's Data Agent Reads 7 Databases. Not One Is on a Factory Floor
- 3 AI Agent Breakouts, 1 Week, and a Red Button From 1991
22 piecesSemiconductors: chips, fabs & yield
Why a chip you can design today still cannot be built for three years.
- Everyone Is Announcing Capacity. Nobody Is Announcing The Number That Decides It.
- AMD Just Bought an AI Chip That Can Never Be Updated
One thing to try on Monday
All 39 tipsGive your AI two lanes, in ten minutes. Open Copilot Cowork, the Microsoft Copilot that does whole tasks for you, and find the reasoning effort control in the prompt box: Faster, or Smarter. Anything that will be acted on goes in the Smarter lane, like a number in a report or an email to a customer, and you add one line to the prompt: "List what you checked and what you could not check." Drafts and brainstorms stay in the Faster lane. Then write the lane at the top of the output so whoever reads it knows which one you used. If you do not have Cowork, ChatGPT and Claude have a thinking mode that does the same; use it on public documents only. A chip factory runs a new machine slow, with extra checks, before it runs it fast. Give your AI the same two speeds.

Twenty years in the software that runs factories. One story a week on what it taught me.
Read about FabSpeak →One email. One story. Once a week.
The week’s breakdown, the one number behind it, the caveat nobody puts in the press release, and one AI trick you can actually use at work on Monday. No roundups. No sponsored sections.
You’re on the list. The first edition lands when it goes out; nothing before, nothing else.
- A correction. The most useful message you can send.
- You work in this layer and something here was wrong, or right for the wrong reason.
- Speaking, podcast or panel.
- Something you want explained. Recurring questions become episodes.