Will AI Replace Jobs? Here’s My Take.

Is AI replacing jobs? Or are we?

Whilst studying a Master’s in Innovation this year. I have regularly spoken with founders, policy leads, VCs, and internal innovation teams across sectors - government, global consultancies, AI incubators. So, this question should feel old by now.

But what keeps it alive is how the answers shift depending on who’s in the room.

Here’s the uncomfortable truth: AI isn’t replacing jobs. People are. And often, they’re doing it before the tools have proven they work.

Take Cisco. They recently cut thousands of roles while posting nearly $54 billion in annual revenue - their second-best year ever. Their CEO earned $32 million. The official line? Resource reallocation for long-term growth. I’ve read their “Responsible AI” principles. It’s well-intentioned. It covers fairness, transparency, governance. But I couldn’t find anything measurable on customer experience. No framework for assessing brand risk. No benchmarks for validating whether the AI tools being deployed were good enough to justify removing the humans.

Same story at Klarna. A 90% reduction in customer service roles, replaced with an AI assistant. And according to them, it’s working - increasing efficiency. But who validated it? Where’s the audit? If you’ve ever been stuck in a chatbot loop while trying to resolve a refund, you already know how “efficient” that can feel on the other end.

So, what’s really happening?

In too many boardrooms, AI is being treated as a savings strategy. A way to shave costs, impress investors, and ride the hype cycle. But that’s not how great innovation works.

In government, I’ve seen something different. i.AI, the UK government's central AI Incubator - are building tools to augment, not replace. Tools that summarise meetings, extract planning data, interpret legislation. Tools that help overstretched teams, not cut them out. And that tells me something important:

This isn’t about what AI can do. It’s about what leaders want it to do.

I’ve spent the last year digging into historical parallels in my academic work. When China ramped up innovation in wind and Electric Vehicle's, it wasn’t just about the tech. It was about infrastructure, re-skilling, and co-ordination. That’s what scaled - not shiny interfaces, but long-term investment in systems.

We’ve seen this before in the .com boom and bust

During the .com boom, companies scrambled to “go digital”. Some built websites. Some raised billions. Most vanished. The ones who survived? Amazon. Google. Salesforce. They built the infrastructure.

So, what’s the infrastructure play in Gen-AI? Is it a chatbot that replaces three support agents and shaves five points off headcount? Or is it the quietly brilliant tooling that enhances every other part of your system?

AI can be transformative. But only if it’s embedded into systems that already work. And only if the people building it know what problem they’re solving.

No, AI won’t replace jobs.

But if leaders keep rushing to do it themselves – without proper impact sizing, no UX validation, no safeguards for trust or service quality - they might just find themselves replaced too.

What do you think? Have you seen AI used to genuinely help people do better work - or replace them before it was ready?