Investment management is contending with an AI skills gap, and the need will only grow more urgent. Firms are adding data scientists, poaching engineers from technology companies, and launching AI centers of excellence. The instinct is understandable: the execution gap between firms that can operationalize AI effectively and those that cannot is real, and it is widening.
But the idea that what investment firms need most is more technical AI expertise misses a more consequential shortage: the skills that determine whether AI deployment actually creates value. Talented individuals in this space live closer to the investment function itself, and most firms have barely started building them.
The framing problem with “the AI skills gap”
When firms describe an AI skills gap, they tend to mean one of two things:
- a shortage of people who can build and maintain AI systems, or
- a shortage of people who know how to use deployed tools
Both are real problems, but neither is the right starting point. The World Economic Forum’s 2025 report on AI in financial services identified a more precise challenge: roughly 90% of financial services leaders believe their organization’s reskilling strategy requires significant adjustment or total transformation to keep up with the pace of AI adoption.
This gap is strategic: a failure to build the organizational capabilities that would make AI investments productive. Firms that attempt to close it by hiring more engineers and data scientists will find themselves better at deploying models, but they may not be able to tell which models matter, what their outputs mean in context, or when a model’s recommendation should be overridden.
The skills firms are underestimating
The CFA Institute’s 2026 analysis of AI skills development is direct on this point: the capabilities investment firms need most are more interpretive than technical. Professionals who can evaluate AI outputs critically have value to firms. To bridge the AI skills gap, teams must be able to identify when model recommendations conflict with market conditions not reflected in training data. They must recognize bias patterns in algorithmic outputs, so it can be corrected.
This distinction matters because of where the real risk concentrates. An AI system that generates a flawed portfolio recommendation is a recoverable problem if a skilled professional catches it. The same system deployed in an organization where familiarity with AI tools is confused with competence in evaluating their outputs creates a more serious exposure. The CFA Institute flags exactly this risk: the widespread use of consumer AI products has created a broad but shallow familiarity with AI that can produce overconfidence in professional settings where the stakes are considerably higher.
The talent function itself reflects the AI skills gap
Here, the AI skills gap becomes recursive. Deloitte’s 2025 research on AI in financial services talent management found that only 18% of financial services firms are currently using generative AI in their talent functions. Many firms that have deployed AI broadly across operations. However, they have not applied the same capability to identifying, developing, and retaining the people who need to govern it.
This creates a compound problem. Not only are the needed skills in short supply, but the systems that would help firms find and build those skills are also underdeveloped.
The implications run through the full talent lifecycle. Succession planning is still largely manual at most firms. Role definitions have not been updated to reflect AI-era expectations.
Many skills frameworks guiding hiring and development decisions were built for a different operating environment and haven’t kept pace. Firms are recruiting against outdated profiles and developing their people for jobs that are less relevant than the ones AI is actually creating.
Firms are stalling on execution
Understanding the problem intellectually has not translated into solving it. Catalyst Partners’ 2026 analysis of talent development in financial services identifies a persistent distance between strategic ambition and organizational delivery. Firms articulate skills-led workforce strategies convincingly in planning documents. The actual work of redesigning roles, establishing new competency frameworks, and building learning programs aligned to AI-era requirements tends to get deferred, deprioritized, or handed to teams without the authority or resources to act on it.
Talent development functions in most investment firms were not built to move at the speed of AI deployment. The AI skills gap grows because skills that mattered in 2022 are being supplanted faster than development programs can respond. But part of it also reflects a broader organizational tendency to treat talent development as a support function rather than a strategic one, something that follows the business rather than shapes it. In the AI era, that sequencing has become a significant liability.
The human skills that become more valuable as automation expands
The Financial Brand’s analysis of the human-AI workforce points toward a useful reframe: as AI absorbs more of the analytical and transactional work in investment operations, certain human capabilities actually become more central.
Critical thinking and AI oversight
The ability to interrogate AI outputs, understand their limitations, and make consequential decisions on the basis of incomplete or imperfect information is a core professional competency for an AI-focused investment firm. This is what regulators are pointing to when they require human accountability for AI-assisted decisions. More than simply being present, a human must be capable of genuine scrutiny.
For example, imagine an investment research workflow where an AI system summarizes earnings calls, analyst notes, and market news into a recommended portfolio action. Its real value comes from a professional recognizing when the model is overweighting stale signals, missing a macro development that changed the investment context, or presenting unwarranted confidence around an uncertain conclusion. Firms need professionals who can evaluate AI outputs critically rather than treat them as inherently authoritative.
Ethical judgment and accountability
Questions about bias, fairness, and harm can be harder to see after a model generates the answer. Professionals who can recognize when an AI-assisted recommendation carries embedded risk that the model has no way to represent, and who are willing to act on that recognition even when doing so slows the workflow, are exercising a capability that no AI system can replicate.
One likely use case is compliance or suitability review. In this case, an AI system flags investor profiles, communications, or portfolio recommendations for potential issues.
A human professional still has to determine whether the model is surfacing a legitimate concern. It could be reproducing bias or overlooking a nuance that matters to the client relationship or regulatory context. As automation expands, accountability concentrates on the people expected to understand when automated judgment is insufficient.
Adaptive communication
AI will take on more and more of the information-processing function. The distinctively human capacity to communicate context, nuance, and uncertainty will be a more concentrated source of professional value.
Relationship quality in financial services is still the primary driver of client retention. AI augmentation actually makes human communication more important. It creates higher expectations for the quality of human interaction when it occurs.
Consider a client-facing scenario in which AI has helped generate portfolio commentary, performance explanations, or responses to a period of market volatility. The operational gain is real, but the decisive moment comes when an advisor, portfolio manager, or client service lead has to translate that information into language that reflects the client’s goals, concerns, and level of confidence.
Building capability rather than chasing it
The firms that are advancing on this are not universally the ones that spent the most on AI talent acquisition. What distinguishes them is that they treat workforce capability development as an operating function. They grant it its own cadence, accountability, and investment thesis rather than simply a complement to AI deployment.
In fact, the firms that prioritize skills-led workforce models have committed to making employee capability information visible across the organization. Their leaders know what skills exist where and where the AI skills gap exists. They track skill development over time and use that information to drive hiring, development, and succession decisions continuously rather than episodically.
This is a capability that most firms have the infrastructure to build. What it requires is treating it as a strategic priority rather than an HR exercise.
The real question for investment management leaders is whether their organizations can field professionals who are AI-competent. They must know what to do with AI, when to trust it, when to override it, and how to explain those decisions to clients and regulators. Firms must build that capability deliberately, as part of how the organization develops over time.
The firms that understand this will build a durable advantage in an environment where the technology itself is increasingly available to everyone.
How Option One Technologies supports investment firms here
Option One Technologies helps build the managed IT and cloud infrastructure that makes capable people more effective. Get secure, scalable infrastructure that allows investment teams to focus more on the judgment-intensive work that creates real value. If you need a platform built for the demands of AI-era investment management, contact a member of our team.
