Stop fearing your AI colleague. Start managing it.

Is AI coming for your job?

I've spent years building AI systems into real businesses, and here's the uncomfortable truth the headlines won't tell you:

AI is not coming for your job. But a person who knows how to work with AI might be.

That's not a threat. That's a design principle. And if you understand it correctly, it's actually the most liberating thing about this moment in history.

The wrong war

The "AI vs. human" narrative is a compelling story. It sells newspapers, drives clicks, and gives talks a tidy villain. But it fundamentally misreads how intelligence, artificial or otherwise, actually functions in organisational settings.

Consider what AI is genuinely exceptional at: processing patterns across enormous datasets, maintaining consistency at scale, never getting tired at 4pm on a Friday, and flagging anomalies a human eye would miss in ten thousand rows of data. Now consider what humans are genuinely exceptional at: reading a room, exercising contextual judgment, building trust, navigating ambiguity, and asking the question nobody thought to ask.

These aren't competing skill sets. They're complementary ones. The tension we feel about AI in the workplace isn't technological, it's organisational. We've inherited work structures designed for human-only labour and we're awkwardly trying to graft AI onto them. No wonder it feels like a fight.

The real question isn't "will AI replace me?" It's "how do I architect my work so that AI handles what it's built for, and I focus on what only I can do?"

Advanced AI Prompt Engineering

Large language models, such as ChatGPT, have transformed how we interact with technology. Leveraging their full capabilities, however, requires more than basic queries. This module delves into the art and science of crafting prompts that elicit...

Learn more

The superpower nobody is teaching

Here's what I see separating the professionals who thrive with AI from those who remain anxious about it: intentionality about the handoff.

Top performers in AI-integrated workplaces aren't the ones who use the most AI tools. They're the ones who've developed a precise mental model of where AI judgment ends and human judgment must begin. They treat AI like a highly capable junior analyst who never sleeps, never complains, and occasionally hallucinates facts with complete confidence, which means they've learned to verify, to push back, and to own the final call.

This is a skill. A learnable, practicable skill. It involves three habits that almost nobody talks about in AI training programmes:

First: give AI context, not just commands. A prompt that says "write a proposal" produces generic output. A prompt that says "write a proposal for a risk-averse CFO who rejected our last pitch because it lacked ROI detail — here's what she pushed back on" produces something actually useful. AI is only as contextually intelligent as the context you give it. The professional who understands this multiplies the tool's effectiveness by an order of magnitude.

Second: use AI to stress-test your thinking, not just produce it. Some of the most valuable AI interactions I've had aren't "write this for me", they're "here's my argument, what are the three strongest counterarguments?" or "what am I missing in this analysis?" AI trained on vast human thought is genuinely useful as an intellectual sparring partner. People who only use it for output generation are leaving the best half of the tool on the table.

Third: keep the relationship visible. In team and client settings, transparency about AI involvement isn't a liability, it's a signal of sophistication. The professionals who say "I used AI to generate the first draft and then stress-tested the logic" are demonstrating judgment, not dependency. The ones who quietly submit AI output as unexamined work are the ones who get burned.

Redesigning the division of labour

At the organisational level, companies that are genuinely winning with AI aren't deploying it to cut headcount, they're using it to redeploy human attention toward higher-leverage work.

A legal team that uses AI to review standard contract clauses doesn't need fewer lawyers; it needs lawyers who spend less time on routine review and more time on strategy and client relationships. A marketing team that uses AI for first-draft copy doesn't need fewer writers; it needs writers who spend less time staring at blank pages and more time on the creative direction that makes the copy worth reading.

The organisations that will struggle aren't the ones with too much AI, they're the ones that never redesign the human role alongside it. When AI handles the scaffolding, humans need to become better architects. That's not a reduction. That's an elevation.

The career move very few are making

If you're reading this as someone navigating your own career in an AI-saturated landscape, here's the most practical advice I can offer:

become the person who makes AI legible to the humans around you.

Not an AI engineer. Not a prompt-engineering specialist. Just someone who can clearly explain what a given AI tool actually does well, where it fails, and how to integrate it into a workflow without losing accountability. This person is extraordinarily rare and extraordinarily valuable in almost every industry right now.

The fear around AI is real, but it's mostly fear of the unknown, fear of being made irrelevant by something you don't understand. The antidote isn't to become an AI skeptic or an AI evangelist. It's to become fluent.

AI isn't your rival. It's your most scalable collaborator, if you're willing to manage the relationship rather than fear it.