Writing is not simply a way of recording thoughts. It is often how thought happens. The same is true of speaking. We begin with an incomplete idea, hear ourselves express it, notice what does not quite make sense, and refine it as we go. That's how our brains work. That is why I worry about using AI too early. If we ask it to produce the first draft before we have formed our own view, it does more than save time. It frames the argument, chooses what matters and sets the direc

Most leaders know what good leadership is supposed to look like. Listen carefully. Stay curious. Involve people. Consider different perspectives. Avoid jumping to conclusions. But knowing what to do is not the same as being able to do it when the situation is ambiguous, the stakes are high, and somebody challenges your first answer. More than 25 years ago, Jeffrey Pfeffer and Robert Sutton gave this problem a name: the knowing-doing gap. Their focus was organisational perform

I came across a fascinating new paper through Dr Philippa Hardman 's excellent newsletter on how AI is quietly reinventing assessment. Researchers at NYU used AI voice agents to conduct oral examinations with students. The AI questioned them about their work, probed their reasoning and asked follow-up questions. The resulting conversations were then assessed against a rubric by several different language models. The study is small and the approach still needs proper validatio

The world has gone a bit mad for AI agents. Claude Co-work, Codex and Copilot agents, … increasingly complex workflows are being handed over to systems built on large language models. Yet somewhere along the way, we seem to have gone strangely quiet about hallucinations and other forms of AI error. In the early days of ChatGPT, if I analysed survey data, it was obvious that deterministic code was being written and executed. The calculations were automatically visible and repr

Based on: Sun, S., Li, Z. A., Foo, M.-D., Zhou, J., & Lu, J. G. (2025). How and for whom using generative AI affects creativity: A field experiment. Journal of Applied Psychology, 110(12), 1561–1573. https://lnkd.in/emUfwNzx

There’s a lot of excitement right now about Claude Cowork. It’s a step towards agentic behaviour: you give it access to part of your computer, and it can automate work across familiar tools like email, spreadsheets and external services. But what are we actually automating? Before we get to value, there’s a background hum of risk we shouldn’t pretend away. Handing AI systems access to your files and workflows assumes nothing goes wrong, nothing hallucinates, and nothing incon

Below is the vison which you can test drive in beta at https://leadershipskillbuilder.ai

The word agent has suddenly become the label for almost anything that moves. A simple workflow is an agent. A chatbot with a memory is an agent. A string of prompts stitched together is an agent. It is the same pattern we saw when agile and disruption slipped into everyday language. Natural enough, but as the meaning spreads, the underlying idea thins out. That thinning matters, because most of what is now called an agent is really an automation. It follows a sequence. It res

A few months ago, a US court had to deal with a surreal problem: a legal brief that confidently cited a series of cases that didn’t exist. The lawyer hadn’t meant to deceive anyone. They’d asked a generative AI model to find relevant precedents and, under pressure, trusted what it produced. The model obliged with fluent nonsense. How does something like that still happen? Probably because the lawyer wasn’t using an integrated system at all. It was almost certainly what we now

My biggest fear about AI Part 2 The influence of the LLMs on what users perceive as true will surpass that of social media. The autocrats already have those models in their sights, and organisations are already complying in advance.