Lawrence Knowles — Blog https://lozknowles.com/blog/ Work, intelligence and responsibility. AI Can Create Value. The Return Still Has to Be Proved. https://lozknowles.com/blog/ai-value-and-return/ https://lozknowles.com/blog/ai-value-and-return/ Infrastructure spending, model capability and customer profit are different measures. What does one accepted outcome cost? Mon, 28 Sep 2026 00:00:00 +0000 The Work AI Leaves Behind https://lozknowles.com/blog/the-work-ai-leaves-behind/ https://lozknowles.com/blog/the-work-ai-leaves-behind/ An agent can finish its part of a task while leaving the organisation with more work to do. Mon, 28 Sep 2026 00:00:00 +0000 Five Technical Problems We Had to Address in Agent Control https://lozknowles.com/blog/five-technical-problems/ https://lozknowles.com/blog/five-technical-problems/ Timeouts, fresh evidence, bounded concurrency, verification and the complete cost of a job. Mon, 28 Sep 2026 00:00:00 +0000 The Agent Is Cheap. The Dependency May Not Be. https://lozknowles.com/blog/the-agent-is-cheap/ https://lozknowles.com/blog/the-agent-is-cheap/ The economics of agents extend beyond the price of a model call. Sun, 27 Sep 2026 00:00:00 +0000 When AI Does the Work, What Happens to Us? https://lozknowles.com/blog/when-ai-does-the-work/ https://lozknowles.com/blog/when-ai-does-the-work/ Jobs, responsibility and who benefits from greater productivity. Fri, 25 Sep 2026 00:00:00 +0000 When AI takes on real work, who remains responsible? https://lozknowles.com/blog/who-remains-responsible/ https://lozknowles.com/blog/who-remains-responsible/ An AI agent can complete a task and still create an operational failure. Fri, 25 Sep 2026 00:00:00 +0000