AI ambition vs. AI reality
The Digital Integration in Wealth Management 2026 conference was an interesting moment to pause and reflect on where we are, right now, with AI.
With everyone seemingly ‘riding the AI wave,’ are we actually using it to transform the wealth industry, or are we just kidding ourselves when the reality is that we’re not even scratching the surface? The answer, like many things, is complicated.
While 97% of wealth management firms globally have a Gen AI initiative underway, investment is forthcoming. But where’s that investment going? Do firms have a strategic plan aligning AI with their growth goals? What do we need AI to actually deliver?
Honest answers about AI investment in wealth
I've spent my career sitting at the intersection of client experience and commercial growth in wealth management. So, when I walked into the conference, I wasn't really expecting to be surprised, but I was wrong.
The room was filled with senior advisors, technologists, compliance professionals, and data specialists from across the industry, pretty much what you’d expect. But what I wasn’t expecting was the depth of honesty they shared. Openly admitting that the gap between their AI ambition and AI reality is wider than most firms are comfortable acknowledging publicly.
Despite the willingness to invest in AI, what that 97% actually tells us is that when every firm in an industry is doing the same thing, it stops being a competitive advantage and becomes a baseline expectation. That’s the pause moment.
The reflection came in the discussions that followed: The firms that will pull ahead aren't the ones with an AI initiative; they're the ones who are clear about why they have one, what they're measuring, and who it's actually for.
Everyone’s investing, but nobody’s transforming
Practical AI implementation is happening. While 89% of wealth management firms describe AI as a core strategic priority, 67% are actively developing or piloting agentic AI. But, when you trace back to see where the real deployment sits, it's advisor efficiency tools, meeting summaries, compliance documentation, and internal knowledge management. Genuinely personalised client engagement and front-office augmentation are still in the early phases, or not even off the bench yet.
So, what’s holding them back?
Bridging the gaping chasm between ambition and reality requires more than investment. It requires clarity about what you’re actually solving, and the honesty to admit what's getting in the way.
Three things came up repeatedly across the sessions I attended, and none of them are new problems. But they're not going away either.
1. The data problem nobody wants to own
Most firms are trying to build AI-ready organisations on top of data that doesn't speak to itself. They’re dealing with multiple legacy systems, inconsistent formats and duplicated records. It’s certainly something I have dealt with over the years; the same client being logged across 3 different platforms. My first thought is: this isn't a failing tech, this is an organisational issue that becomes glaringly obvious when you try to do anything meaningful with AI.
That thought was shared by the room and the stats: data quality was ranked as the single biggest barrier to AI implementation across the research presented at the conference. Not budget, regulation or even capability. It was the foundation of the firm's data management.
One of the most useful points made was this: the goal isn't to send your data up to the model, it's to bring the model down to the data. Keeping it within your own infrastructure and governance controls matters because it shifts the conversation from ‘which AI tool do we use?’ to ‘is our data architecture actually ready for this?’
Until firms are honest about the state of their data, AI will continue to deliver exactly what it does today: useful but limited change. And while that’s impressive in isolation, it will prove underwhelming at scale.
2. The regulatory gap is real but shouldn’t be an excuse
The FCA's position on AI remains deliberately high-level and light-touch. The tech is new and too rudimentary to embed in operational policies and procedures with any confidence, and that gap is real and risky. Especially for back- and middle-office functions, the absence of clear guidance creates genuine hesitation.
But what I witnessed in the room was a range of responses to that ambiguity. Some firms are treating it as permission to move, opening options, building carefully and documenting their decisions. Others are using it as a reason to wait, cautious about capabilities, costs and security. And critically, there's still no industry-wide best practice. Nobody who says: here are the use cases that are working, here's what good looks like, here's the standard we're holding ourselves to. That gap is costing the industry more than the regulatory uncertainty itself.
3. The capability gap is cultural, not just technical
Perhaps one of the most candid observations came when someone pointed out that they hadn’t even started using AI tools at work. Not because of fear or avoidance, but simply because they hadn’t been given a compelling reason to change, and no one had made it easy enough to start. I have to say that it did get some wry smiles from others in the room.
Because the training had been mostly a list of what not to do rather than what it could help them achieve, they hadn’t been persuaded to even try. Now, that's a culture and change management problem, and it's one that no amount of vendor investment will solve on its own.
Freeing up time is step one, but then what?
This is the question I keep bringing back to firms, and the one I think matters most heading into the second half of 2026.
If AI genuinely does what it promises (clawing back the 50-60% of advisor time currently lost to documentation, administration, and process overhead), what are advisors doing with that new regained capacity?
Having spent years in wealth management businesses, I've seen this cycle before. Efficiency improves, but without a strategic plan for the best use of that time, it gets absorbed back into the existing workflow rather than redirected toward something more valuable.
The competitive firms will be the ones who ask, early and seriously: we're about to free up significant advisor time. What do we want them to be doing with it? And then build the answer into their strategy before the capacity arrives.
The industry's no.1 priority is retention, not acquisition.
That might mean a deliberate expansion of the client book into segments that weren't commercially viable before. It might mean a deeper, more proactive service proposition for existing clients, the kind that builds the trust no platform can replicate.
We know baby boomers, millennials and Gen Z think differently about wealth and investments. McKinsey predicts that eight in ten firms will focus exclusively on retention over the next twelve months, because clients are nowhere near as sticky as they were even three years ago. That should shape marketing strategies, not just service delivery.
It might mean investing in intergenerational relationships that will determine whether wealth remains with the firm upon transfer. None of those answers is wrong. But they are choices, and they need to be made intentionally.
The firms that will pull ahead
The most useful reframe I took from the conference was this: stop asking how to become AI-first and start asking how to become AI-conducive.
For large incumbents carrying decades of legacy, AI-first is an aspiration that can quickly become an expensive distraction. AI-conducive is a practical question: do we have the data foundations, the governance, the cultural readiness, and the strategic clarity to absorb what AI can actually do?
The 97% figure tells us the industry has decided AI matters. What it doesn't tell us is whether firms have done the harder work of deciding what they need it to deliver, for whom, and how they'll know when it's working.
That's the conversation that separates firms with an AI initiative from those building something that will actually last.