# Newsletter > The CAIO: my monthly newsletter for CTOs, CIOs and AI leaders on building organisations that can actually run AI in production. Source: https://romanoroth.com/en/newsletter/ **The CAIO** is my monthly newsletter for CTOs, CIOs and everyone who has to make AI work inside a real organisation. Each edition takes the patterns I see in the field and turns them into something you can act on: what is working, what is breaking, and what it means for how you build and run technology. Every edition is published here in full, in English and German. New editions reach subscribers first, so [subscribe on Substack](https://romanoroth.substack.com/) if you want them in your inbox. Editions 1 to 4 went out under the newsletter's original name, The Cybernetic CxO. The rename is explained in Edition 5. ## In this section (10) - [Cut the Jobs and Hope AI Fills the Gap](https://romanoroth.com/en/newsletter/e10-cut-the-jobs-and-hope-ai-fills-the-gap/), 2026-10-25: Edition 10. Sunrise cut jobs and nobody deleted a step. Why a coding assistant on the old process is a bigger motor in the old factory, what makes human in control measurable, why an agent has the access someone gave it, and who owns what the organisation knows. - [Your AI Bill Can Sink You](https://romanoroth.com/en/newsletter/e09-your-ai-bill-can-sink-you/), 2026-09-27: Edition 9. Token prices halved and the AI bill did not. Why someone has to own the switch between models, why thirty copies of one model are not a second opinion, and what happens to the hours AI gives back. - [It Does Not Matter Who Typed It: When Anyone Can Ship, the Gate Is the Job](https://romanoroth.com/en/newsletter/e08-it-does-not-matter-who-typed-it-when-anyone-can-ship-the/), 2026-08-23: Edition 8. When anyone can ship, the gate becomes the job. Why a green build is not evidence, why your agent is a distributed system, and why nobody notices a skill that was never built. - [The Bottleneck Was Never the Model: The Month the Discipline Argument Got Its Numbers](https://romanoroth.com/en/newsletter/e07-the-bottleneck-was-never-the-model-the-month-the-discipline/), 2026-07-19: Edition 7. The month the discipline argument got its numbers. Why the bottleneck was never the model, why 596k is not the scary number, and how AI amplifies whatever an organisation already is. - [The Bill Arrives: Is This AI Request Worth 100 Dollars?](https://romanoroth.com/en/newsletter/e06-the-bill-arrives-is-this-ai-request-worth-100-dollars/), 2026-06-27: Edition 6. The AI bill arrives and the question becomes whether a single request is worth 100 dollars. Sovereignty as an architecture decision, and what it means that adoption keeps rising while trust falls. - [Afraid of Your Own Culture](https://romanoroth.com/en/newsletter/e05-afraid-of-your-own-culture/), 2026-05-24: Edition 5, and the first issue under the new name The CAIO. What 500 senior leaders admitted about their own culture, why the harness around the model is the product, and what the lump of labour fallacy looks like a hundred years on. - [Democratization of Liability: When Vibe Coding Hits the Real World](https://romanoroth.com/en/newsletter/e04-democratization-of-liability-when-vibe-coding-hits-the-real/), 2026-04-25: Edition 4. When anyone can ship code, liability spreads with it. Why an exoskeleton is a better design goal than an AI coworker, and why discipline beats intelligence once you look at the ROI numbers. - [The 750 Billion Dollar Question: Is AI Actually Replacing Anyone?](https://romanoroth.com/en/newsletter/e03-the-750-billion-dollar-question-is-ai-actually-replacing/), 2026-03-22: Edition 3. The 750 billion dollar question behind the AI jobs debate. What 2026 actually revealed about AI and developers, and the governance gap that opens as soon as agents act on their own. - [Why 93% of your AI budget is in the wrong place](https://romanoroth.com/en/newsletter/e02-why-93-of-your-ai-budget-is-in-the-wrong-place/), 2026-02-22: Edition 2. Why 93 percent of your AI budget sits in the wrong place, eight predictions for 2026, and what agentic islands mean for teams that expect autonomy today. - [Why Moving Fast With AI Often Means Moving Backward](https://romanoroth.com/en/newsletter/e01-why-moving-fast-with-ai-often-means-moving-backward/), 2026-01-26: Edition 1. Teams ship AI-generated code that nobody understands and call it progress. Why AI gets you to 90 percent fast and the last 10 percent eats the year, what happens when agents start replacing SaaS, and why Europe's AI question moved from best model to trusted infrastructure.