This is a personal perspective from Neil Holden, our Co-Founder, shaped by the conversations he’s had over the past year with boards, CEOs and leadership teams across multiple sectors, and by the work we’re doing day-to-day helping organisations turn AI ambition into real outcomes.
As we head into 2026, I’ve been reflecting on the conversations I’ve had over the last year with boards, CEOs, CIOs, CFOs and transformation leaders across retail, manufacturing, logistics, housing, financial services and the public sector.
Different sectors. Different challenges. But a surprisingly consistent pattern.
Most organisations now believe AI matters.
Far fewer are confident they’re doing the right things with it.
And even fewer are translating it into sustained commercial value.
After more than two decades in senior leadership roles inside large organisations, and the last several years building and running an AI consultancy working at the coalface of delivery, my view is this:
2026 will be the year that separates AI tourists from AI operators.
Not in terms of ambition.
Not in terms of spend.
But in terms of execution, discipline, and leadership maturity.
The AI Conversation Has Shifted – But Not Enough
In 2023 and 2024, most AI conversations were exploratory.
In 2025, they became more urgent.
By the end of this year, nearly 75% of CEOs globally are expected to say AI materially impacts their business model (PwC). Yet fewer than 30% report seeing measurable ROI from their AI investments (McKinsey).
That gap isn’t caused by technology immaturity.
It’s caused by organisational behaviour.
The biggest mistake I still see is treating AI as:
- A technology upgrade
- An IT initiative
- Or worse, an innovation experiment parked at the edge of the business
AI is none of those.
AI is an operating model change. And in 2026, businesses that fail to treat it that way will quietly fall behind those that do.
Data Is Still the Constraint – But Leadership Is the Bigger One
We’ve all heard “AI is only as good as your data”. It’s true. But it’s no longer the most useful insight.
What I see repeatedly is this:
- Data quality is known to be poor
- Ownership is unclear
- Accountability is diluted
- And everyone is waiting for “the platform” to fix it
In reality, data problems persist because leadership avoids the hard calls.
Who owns customer data end-to-end?
Who is accountable for master data quality?
Who decides what not to prioritise?
The organisations making progress in 2025 weren’t the ones with perfect data. They were the ones willing to:
- Define “good enough” pragmatically
- Fix data in service of outcomes, not architecture
- And hold senior leaders accountable for data domains
By 2026, data strategy without clear executive ownership will be a red flag, not a maturity signal.
AI Will Expose Weak Operating Models
One uncomfortable truth: AI doesn’t just automate inefficiency – it amplifies it.
I’ve seen organisations deploy AI into:
- Broken processes
- Ambiguous decision rights
- Poorly defined roles
- And fragmented incentives
The result? Disappointing outcomes, mistrust in the tech, and a quiet rollback.
The businesses succeeding are doing the opposite:
- Simplifying processes before automating
- Reducing handoffs
- Clarifying decision logic
- And embedding AI where humans already struggle with volume, speed, or complexity
In 2026, AI will increasingly be a mirror. It will show leaders where their operating model is unclear – and where they’ve been avoiding change.
The Talent Question Is Being Asked the Wrong Way
There’s still far too much focus on:
“Do we need more data scientists?”
“Should we hire prompt engineers?”
“How do we compete for scarce AI talent?”
In my experience, that’s not where the leverage is.
The organisations pulling ahead are investing in:
- AI-literate leaders, not just specialists
- Upskilling domain experts, not replacing them
- Teaching teams how to work with AI, not around it
Microsoft research suggests AI-augmented knowledge workers can see productivity uplifts of 20-40% when deployed well. But only when the organisation redesigns work, not just tools.
In 2026, competitive advantage won’t come from who has the best models.
It will come from who redesigned roles, incentives, and decision-making fastest.
Governance Will Move From Risk Control to Value Enablement
For many boards, AI governance has understandably been framed around risk:
- Ethics
- Bias
- Data privacy
- Regulation
Those things matter. Hugely.
But the next shift I’m seeing – and strongly agree with – is governance evolving into a value-enabling function, not a blocker.
The best boards I work with are now asking:
- Which decisions should be augmented by AI?
- Where are humans still essential – and why?
- What level of accuracy is “safe enough” for different use cases?
- How do we monitor drift, not just deployment?
By 2026, boards that only ask “is this safe?” will lag behind those also asking “is this transformative?”
AI Strategy Must Be Boring to Work
This might sound counterintuitive, but it’s one of my strongest beliefs.
If your AI strategy is exciting, flashy, or full of moonshots – it’s probably wrong.
The most effective AI programmes I’ve seen are:
- Ruthlessly prioritised
- Tied to P&L or service outcomes
- Incremental, not heroic
- And frankly… a bit boring
They focus on things like:
- Reducing rework
- Improving forecast accuracy
- Speeding up cycle times
- Eliminating low-value manual effort
These don’t make headlines.
But they compound.
And by 2026, compounding advantage will matter more than innovation theatre.
What I’d Be Asking as a CEO Going Into 2026
If I were sitting in a CEO or board seat today, these are the questions I’d want clear answers to:
- Where exactly are we using AI today – and where are we pretending?
- Which business outcomes improved last year because of AI?
- Who owns AI value delivery – not experimentation?
- What decisions are still slow, manual, or inconsistent – and why?
- Do our leaders understand AI well enough to lead with it?
If those answers are vague, 2026 will be uncomfortable.
A Final Thought
AI is no longer a future capability. It’s a present-day leadership test.
The winners won’t be those who adopt the most tools, shout the loudest, or spend the most. They’ll be the organisations that:
- Take AI seriously enough to be pragmatic
- Treat it as an operating shift, not a tech project
- And lead it with clarity, discipline, and intent
That’s where we’re focused at Ignite AI Partners – helping organisations move from curiosity to capability, and from pilots to performance.
2026 isn’t about whether AI will change your business.
It’s about whether your leadership team is ready for that reality.
