All conversations about AI in boardrooms now follow a similar script. The investment project will be completed. The tool is selected. The expansion has been announced. And somewhere between pilot and production, things slow down in a way no one had planned.
Technology is rarely the culprit.
In my nearly 20 years of experience in technology recruiting and workforce strategy, I’ve seen this pattern repeat itself across industries and company sizes. Organizations struggling to scale AI are rarely doing so because their models aren’t good enough or their platforms aren’t mature enough. They struggle because they don’t have the talent to run, manage, validate, or integrate these systems at any meaningful scale.
This is a workforce bottleneck that enterprise AI conversations consistently don’t focus on, and it’s a real and measurable drag on the bottom line.
The numbers behind the gap
The World Economic Forum’s Future of Work Report 2025 identifies the skills gap as the single biggest barrier to business transformation, with 63% of employers citing it as a key constraint by 2030. Separately, while 85% of these same employers plan to prioritize upskilling, only 32% of companies are confident that their company already has the skills needed for long-term success.
The gap between intent and readiness is where the ROI of AI quietly disappears.
According to 2025 data from Nash Squared, demand for AI skills has nearly doubled year over year, jumping from 28 percent of companies reporting AI as a priority skill in 2024 to 51 percent in 2025. In the United States alone, the number of jobs requiring AI proficiency has increased by more than 1,800 percent in the last two years. Supply is not keeping up. In financial services and healthcare, two of the industries that are investing the most in enterprise AI, the average time to hire for AI-related roles is six to seven months.
6-7 months. It is often a central role in executing the technology strategy an organization is already working on.
What is actually needed for “AI readiness”
One of the problems is that the response to AI is very narrowly framed. Companies look for machine learning engineers and data scientists, but it’s really hard to find them, and they stop there. The range of skills is quite wide.
Effective enterprise AI implementation requires people who can do at least six different things. This means designing and maintaining data pipelines that feed into models, translating business requirements into AI-friendly problem frames, assessing accuracy and bias of model outputs, managing compliance and auditability in regulated workflows, communicating system behavior to non-technical stakeholders, and handling change management as workflows transform.
Most organizations staff one or two of these and assume the rest will take care of itself. it’s not.
According to the Workday 2025 Global State of Skills survey, only 54% of business leaders say they have a clear understanding of the skills of their existing workforce. You can’t make a reasonable plan for what you need if you don’t know what you already have. And you can’t build a company fast enough if a six-month vacancy destroys the ROI projections you laid out in front of the board.
ASEAN-specific aspects
For companies operating across Southeast Asia, there are many more layers to this problem.
The regional talent distribution for AI skills is very uneven. It is concentrated in some metropolitan areas of Singapore, parts of Malaysia, and Indonesia, but depth decreases rapidly outside these hubs. The localization requirements for ASEAN language processing add an additional layer of specialization that is typically not covered by global AI vendor solutions. A model trained primarily on English data requires meaningful local adaptation to work reliably in Thai, Bahasa, Vietnamese, or Tagalog-enabled applications.
This adaptation requires talent with both the technical depth to work at the model level and the local market knowledge to meaningfully evaluate the output. This particular combination is currently one of the rarest profiles in the region’s talent market.
Companies tackling this issue are doing so through intentional strategies, such as building partnerships with universities with strong domestic AI programs, investing in the systematic upskilling of subject matter experts who already understand the local market, and working with workforce partners who maintain pre-vetted talent pipelines rather than starting the search from scratch when an urgent need arises.
The answer to internal mobility is one that no one is taking seriously enough
Before expanding your search externally, the most cost-effective first step is often to take a closer look at what already exists internally.
According to SHRM’s 2025 data, internal talent market adoption increased from 25 percent to 35 percent of organizations in one year. This growth reflects a real change in understanding. The people best suited to support AI deployment are often those who already know the business processes to which the models are being applied. A claims analyst who understands edge cases in underwriting is a more valuable AI validation resource than a generalist data scientist who has never seen the field.
Preparing that talent for a new role requires investing in systematic upskilling, rather than a six-month external search. Economics is straightforward. Barriers are usually organizational, specifically the belief that internal candidates are not technically good enough without actually being evaluated.
What makes high-performing organizations different?
Companies that scale AI effectively share several operational patterns that are worth directly naming.
AI’s role is defined by what it actually needs to do the work, not a credential proxy. A job description that requires a computer science degree and five years of machine learning experience will filter out a large portion of the people who can do the job well.
They treat talent pipeline development as part of their AI strategy rather than downstream HR operations. The best time to start building a bench that accommodates the functionality you need is before it becomes an emergency.
They work with external partners who understand AI skill requirements at a granular level, not just at the role level. The difference between someone who has “used AI tools” and someone who can build production-ready inference pipelines is hard to tell on a resume without much digging.
And they measure the right things. Time to fill AI roles, quality of adoption at six months, and gaps between projected and actual AI delivery timelines are all connected to data points. Organizations that track these together are more likely to discover and address the impact of talent on AI ROI earlier.
practical final conclusion
The reason AI is stagnant is not because the technology is immature. In most enterprise environments, the platform has greater capabilities than the organization that deploys it.
Humans are the constraint. Analysts are the bridge between model output and business decisions, engineers are the ones who can maintain the pipeline at production scale, and compliance officers understand what auditability means in specific regulated situations.
Finding, developing, and retaining that talent is what determines whether an investment in AI will deliver the promised results. Organizations that treat the job as strategic, rather than as a sourcing issue to hand over when a role becomes available, will make much more headway than those that don’t.
The technology gap has closed faster than most expected. Next is the talent gap. The companies that finish it first will gain lasting benefits that are harder to replicate than the tool itself.
milind naik He is Vice President of Sales and Recruiting for Compunnel Inc. and has over 19 years of experience in customer recruitment in the United States. His focus is on identifying high-impact technology talent and scaling recruiting teams to meet complex hiring demands. With deep expertise in workforce growth and customer engagement, Milind shares his first-hand, experienced perspective on Compunnel Inc.’s recruiting strategies, talent dynamics, and workforce solutions.
TN Global Insider Publish contributions related to entrepreneurship and innovation. You may submit your own original or published submissions, subject to editorial discretion.
Featured image: Campaign Creator on Unsplash
Keeping pace with industry disruption in the AI era: Four strategic imperatives
