I recently joined Jon Kaye, on DO Radio, for a conversation that went far beyond just technology. We got into agentic AI, the future of work, and what really happens when workflows start to optimise themselves. But more importantly, we talked about people - how we learn, how we adapt, and how we make sure technology works for us, not the other way around. What we covered… Agentic AI - what it actually means We started by unpacking what “agentic AI” really is. At its core, it’s about breaking work into steps and deciding what should be done by a human and what should be done by a machine. That’s not new in itself - we’ve always had workflows - but what’s changing is how much of that can now be handled by intelligent systems. For me, the key point is this: it’s not about replacing humans. It’s about designing better human + machine systems. The reality of AI in business today There’s a lot of noise in the market right now. At one end, you’ve got companies spending billions building proprietary tools. At the other, people are wrapping ChatGPT and calling it innovation. The interesting work sits in the middle. Most of the businesses I work with are using accessible tools - but they’re combining them in smarter ways. That’s where the real gains are happening. In particular, I’m seeing content businesses break the historic link between growth and headcount. They’re scaling output while holding - or even reducing - team size. The best ones aren’t cutting people aggressively, they’re evolving roles and moving people into higher-value work. Tools, workflows, and experimentation A big part of the conversation was about how accessible this all is. You don’t need to build your own models. You can get a long way with tools like ChatGPT, Claude, or Perplexity. The shift comes when you start connecting them together. For me, one of the most powerful tools has been n8n - a workflow orchestration platform that lets you build end-to-end processes. I’ve effectively created a small “team” of agents that: * Capture and transcribe my calls * Turn them into structured actions * Organise everything by client * Timebox my follow-ups * Nudge me to actually get things done That’s not about replacing people. It’s about unlocking capability that wouldn’t be commercially viable to hire for. The broader point is simple: set aside time each week to experiment. You can’t break this stuff. You just have to start. The changing shape of work This is where it gets more complex. There’s a lot of optimism around AI augmenting senior roles - and rightly so. But there’s a real question about what happens to junior roles. Historically, a lot of learning came from doing repetitive, lower-value work. That’s how people built judgement. If we remove that layer entirely, we need to rethink how people develop. It’s one of the reasons I’m increasingly interested in apprenticeships and on-the-job learning. The traditional “university as default” model is starting to look less convincing in a world that’s changing this quickly. Productivity gains - and what we do with them One of the more interesting moments in the conversation was that if AI gives you back ten hours a week, what do you do with it? I shared a story about someone I mentor who did exactly that - and then spent most of the time scrolling on their phone. That can’t be the outcome. The real opportunity isn’t just efficiency. It’s reinvestment - in learning, creativity, health, or building something new. We haven’t really figured this part out yet. Technology, inequality, and responsibility My biggest hope is that this wave of technology flattens opportunity. That it allows more people - regardless of background - to do more, earn more, and create more. My biggest concern is the opposite. If access, understanding, and adoption are uneven, we risk widening the gap between those who can leverage these tools and those who can’t. There are also big unanswered questions around regulation. The people building these systems shouldn’t be the only ones shaping the rules. But equally, we haven’t yet found the right alternative. My work in private equity We also touched on what I actually do day-to-day. My work tends to fall into three areas: * Pre-deal - assessing whether a business can use product and technology to scale * Value creation - helping portfolio companies grow through new products and capabilities * Transformation - moving businesses from manual or analogue processes into more scalable, automated models It’s a fast-paced environment, but what I like about private equity is the clarity. Everyone is aligned on the goal: build value, and do it quickly. Why I write We finished on writing. For me, writing serves a few purposes: * It replaces the “water cooler” conversations I don’t get working independently * It helps me think and remember - once I’ve written something, it sticks * It brings everything I’m doing into one place * And importantly, it gives me a creative outlet alongside more structured work I don’t write because I have to. I write because it helps me make sense of things. Final thought This wasn’t really a conversation about AI. It was a conversation about how we adapt to a world where the rules are changing quickly - and where the biggest risk isn’t the technology itself, but how we choose to use it. Get full access to Chiefly Product at chieflyproduct.substack.com/subscribe
