The job titles that prove AI's hard part was never the technology
By Polly Barnfield, OBE, CEO of Maybe*
Every year, Computer Weekly publishes its Most Influential Women in UK Tech list. It is read, understandably, as a story about diversity: how many women are being recognised, and whether that number is moving in the right direction.
But this year the list contains a second story, hidden in plain sight, that has almost nothing to do with diversity and everything to do with how AI is actually being adopted across UK organisations. It is written not in the names, but in the job titles. And it is one of the clearest pieces of evidence we have seen for something we have argued for two years: the hard part of AI was never the technology.
I have been named on this list every year for the past decade, which gives an unusually long vantage point on how it changes. Here is what the 2026 list reveals when you read it as a map of AI adoption.
The number of AI roles almost doubled in a year
Set the names aside and read only the roles. Count the entries with AI explicitly in the job title, and compare year to year.
On the 2025 longlist, there were eighteen. On the 2026 longlist, thirty-three. In a single year, on the same list, the number of senior women whose actual job is AI almost doubled.
That is a signal in itself. Organisations across the UK are creating dedicated AI roles at a pace that was not visible even twelve months ago. But the raw number is the least interesting part of the finding.
The roles that are growing are about embedding, not building
When you look at what those AI roles actually are, a striking pattern appears. They are overwhelmingly not the roles you might expect. They are not, in the main, the people building the models.
The titles that recur are these: head of GenAI enablement, director of AI adoption, chief AI transformation officer, chief AI readiness officer, AI governance lead, responsible AI, AI assurance, AI ethics. Roles about embedding AI into an organisation, adopting it, governing it, and being accountable for it.
We compared roles focused on embedding and governing AI with roles focused on building the technology itself. The ratio was roughly ten to one. Embedding and governance roles made up around six in ten of all the AI roles on the list. Hands-on technical builders were a small minority.
The conclusion is hard to avoid. The AI roles multiplying fastest are not about making the technology more capable. They are about making it actually work inside a business.
Why job titles are honest evidence
Job titles are a more honest signal than almost anything an organisation says about AI, because they cost money. A press release about an AI strategy is cheap. A Chief AI Transformation Officer, with a salary and a mandate, is a statement of belief backed by budget.
So when organisations create senior roles for AI adoption, AI enablement and AI governance, and comparatively few for building models, they are telling you exactly where they now believe the difficulty lies. Organisations do not create senior roles for problems they consider easy. The rise of the AI adoption and AI governance role is a market-wide admission that the challenge has moved from can the technology do this to how do we get it to reliably do this, safely, inside how we already work.
That is the central finding of our own research into AI in business, arrived at independently, from the opposite direction. Adoption is easy and near universal. Value is rare. And the difference between the two is not the model. It is whether an organisation changes how the work actually gets done, and can trust and govern the result. The job titles on this list are that argument, made visible in other people's org charts.
What this means for how you adopt AI
If the organisations investing most seriously in AI are hiring for embedding and governance rather than raw technical capability, there is a lesson for any business deciding how to approach AI.
It suggests the question isn't which model or tool. Everyone has access to broadly the same technology. The question is how a specific piece of work will actually change, who will be accountable for the result, and how you will know it is right. That is an execution question and a governance question, not a technology question. It is also, notably, the question the most advanced organisations are now hiring dedicated people to answer.
The practical implication is not to write a sweeping AI strategy or to buy another tool. It is to take one real, well-defined piece of work, change how it is done using AI, make sure a person remains accountable for the output, and measure what actually changes. Then do it again. That is the discipline the new wave of AI adoption and governance roles exists to bring, and it is available to any organisation willing to apply it, with or without the job title.
The two stories are one story
There is a final connection worth drawing, because it ties the AI finding back to the list's original purpose.
The AI roles growing fastest depend on judgement, context, and an understanding of how people and organisations really work: whether a decision can be trusted, whether an output is right, whether an organisation is ready. These are precisely the human, judgement-heavy capabilities that a narrow view of technology has always undervalued.
As the centre of gravity in AI shifts from building models to embedding them wisely, the skills that matter most shift too, towards judgement and accountability. That shift ought to widen the field, not narrow it. It also carries a real risk: AI is being built on data that reflects existing biases, and if the people embedding and governing it are not drawn from a wide enough pool, today's imbalances will be encoded into tomorrow's systems. Getting a genuinely diverse set of people into building and governing AI, not just using it, has never mattered more.
Everyone is adopting AI. Few have cracked how to get value from it. The organisations that will are already visible, on a list, hiring exactly the people who understand that the technology was never the hard part.
