Build vs. Buy: Why Internal Development Beats External Hiring Right Now

The external AI talent market is thin, expensive, and unreliable. For most enterprise organizations, the better path is developing the talent already inside the organization.

The argument for external AI hiring is intuitive: you need capability quickly, the market has candidates who already have it, and hiring is faster than training. The problem is that each of these premises is weaker than it appears. The market does not have as many qualified candidates as the pipeline suggests. The time-to-productivity for an external AI hire in a new organizational context is often longer than anticipated. And building genuine AI capability externally does not solve the retention problem that comes after it.

The WEF Future of Jobs Report 2025 found that a majority of employers surveyed planned to upskill existing workers rather than hire externally to fill AI-related roles. This preference reflects a practical calculation: the domain knowledge already inside the organization is often the scarce ingredient, and it is not available in the external market at any price.

The External Market Problem

The external AI talent market has three structural problems that are unlikely to resolve in the near term.

Thin supply at the top. As argued in Part 1 of this series, the fraction of people with AI access who have developed genuine mastery is small. The large candidate pipeline that many organizations see is composed mostly of people with tool familiarity and resume optimization, not people with the cognitive restructuring and compounding leverage that makes AI investment actually pay off. Hiring from that pipeline without a calibrated filter produces hires who look qualified and underdeliver.

Inflated expectations on both sides. Candidates in the current market have often absorbed inflated compensation expectations from a period of intense competition for AI talent that has not fully corrected. Organizations have often absorbed inflated expectations about how quickly an external AI hire will generate returns. Neither set of expectations is grounded in a clear understanding of what high-quality AI work actually requires in an enterprise context, which takes time to develop regardless of the candidate's previous experience.

High churn in the absence of the right conditions. Part 6 of this series addresses the organizational conditions that AI talent needs to succeed. The point here is that without those conditions, external AI hires churn at high rates, typically within twelve to eighteen months. The cost of a failed AI hire, including recruiting fees, onboarding time, lost momentum, and the organizational skepticism that follows, is substantial. Organizations that hire externally without first building the organizational conditions are solving the wrong problem.

The domain knowledge that makes AI applications valuable inside an enterprise is already inside that enterprise. It is not available in the external market at any price.

Why Internal Development Works Better

The case for internal development rests on three structural advantages that the external hiring path cannot replicate.

Domain context is the scarce ingredient. The McKinsey State of AI 2024 report found that organizations reporting the highest AI value were more likely to have AI capabilities embedded in core business functions rather than isolated in standalone teams, suggesting that domain integration is a consistent differentiator. An external AI hire, regardless of their previous experience, arrives without the domain context of your organization: its customers, its processes, its regulatory environment, its failure modes, and its institutional history. A tenured employee who has developed genuine AI fluency already has that context. The combination of deep domain knowledge and AI fluency, developed inside the organization on actual problems, is more valuable and more durable than either element alone.

AI fluency can be taught with the right structure. The cognitive habits that distinguish top AI practitioners from occasional users, described in Part 3 of this series, are learnable. They are not innate traits. They develop through deliberate, sustained practice on real problems with real stakes, supported by the right feedback structure. This is exactly what a well-designed internal development program can provide, and exactly what a weekend certification course or a one-day workshop cannot.

Retention and loyalty follow from investment. Organizations that invest meaningfully in the development of their existing people create retention dynamics that external hiring cannot replicate. People who have been developed are more likely to stay, more likely to develop others in turn, and more likely to identify opportunities for AI application in their domain because they have the organizational knowledge to see where the leverage is. The return on internal development compounds over time in ways that external hiring does not.

What a Real Internal AI Fluency Program Looks Like

The Role of the CHRO in Internal Development

Internal AI development programs often fail not because of poor program design but because of weak organizational ownership. The most successful programs have a senior HR leader, typically the CHRO or VP of Talent, who treats the program as a strategic workforce investment rather than a learning and development initiative. The distinction matters because it changes how the program is resourced, how participants are selected, how outcomes are measured, and how the program is positioned internally.

An internal AI fluency program that is positioned as a learning initiative gets measured on completion rates, participant satisfaction scores, and assessment performance. An internal AI fluency program that is positioned as a strategic workforce investment gets measured on the before/after workflow changes that participants produce, the business value of those changes, and the retention rate of participants compared to the broader employee population. The second measurement framework produces a program that is taken seriously by the business and produces evidence of value that justifies continued investment. The first produces a program that is optional, deprioritized, and ultimately not transformative.

The WEF Future of Jobs Report 2025 found that organizations expecting the highest returns from AI investment were also the organizations with the most structured approaches to workforce development. The correlation is not causal in a simple sense: better-managed organizations tend to do both. But the pattern is consistent enough to suggest that the organizations most likely to capture AI value are the ones that treat workforce development with the same rigor they apply to technology investment.

When External Hiring Still Makes Sense

Internal development is the better default strategy for most enterprise organizations at most stages of AI maturity. But there are specific situations where external hiring is the right call.

Founding team formation. If the organization is building a new AI capability from scratch and has no one internally who has done this before, a small number of external hires with relevant experience can anchor the program: not to fill the entire team, but to bring the institutional pattern-recognition that comes from having built something similar elsewhere. This role is different from a standard AI hire: it is a builder role that requires experience with the organizational challenges of standing up AI capability, not just the technical elements.

Highly specialized technical roles. There are specific technical roles, particularly around model development, infrastructure at scale, and certain research-adjacent functions, where the relevant expertise is genuinely scarce and where the internal development path is impractical because no one inside the organization has the technical foundation to build from. In these cases, external hiring for the specific technical role is appropriate, paired with an internal development strategy for the broader AI fluency population.

The error most organizations make is treating external hiring as the primary strategy and internal development as a secondary or supplementary effort. Given the structural characteristics of the external market and the structural advantages of internal development, these priorities should usually be reversed. A combined strategy that uses external hiring for the founding role and specific technical gaps, while building the broader AI fluency population internally, captures the advantages of both approaches without over-relying on either.

What Internal Development Actually Costs

A common objection to the internal development argument is that it is slower and more expensive than hiring. The cost comparison is worth examining carefully because it is often made without accounting for the full cost of either path.

External hiring costs include recruiter fees, which for senior AI roles typically run 20 to 30 percent of first-year compensation. They include the time cost of interviewing, which for a senior hire often involves 10 to 20 hours of internal leadership time spread across multiple rounds. They include onboarding time, during which the hire is operating at reduced capacity while learning the organization's context. They include the productivity drag that comes from domain context acquisition, which is typically measured in months for a genuinely complex enterprise context. And they include the churn cost if the hire does not work out, which in addition to the direct costs requires restarting the process and includes the organizational skepticism that follows a failed senior hire.

Internal development costs include program design and facilitation, which for a well-structured cohort program is a real investment. They include the time cost of participants, who are investing their own time in the program while continuing their other work. They include the opportunity cost of deploying people on development activities rather than their current roles.

The comparison shifts significantly when the full cost of each path is considered, especially when the churn cost of a failed external hire is included. The McKinsey State of AI 2024 report found that high-performing AI organizations were more likely to embed AI capabilities within business functions rather than keep them separate. The implication is that the combination of domain knowledge and AI fluency is what produces value, and internal development is a more reliable path to that combination than external sourcing.

The Compounding Argument for Internal Development

The strongest argument for internal development is not the cost comparison on any individual hire. It is the compounding dynamic that emerges when internal development is treated as an ongoing capability rather than a one-time event.

An externally hired AI specialist brings their capability in and applies it. An internally developed AI practitioner brings their capability in, applies it, and then in most cases becomes a resource for developing others. They already know the organizational context, the domain, the processes, and the people. They can teach what they know in ways that are immediately applicable to the organization's actual problems, not generic principles that require translation. The return on the development investment extends beyond the individual to whoever they influence, mentor, and enable.

This compounding dynamic does not happen automatically. It requires an organizational structure that recognizes and rewards knowledge transfer, that creates the conditions for practitioners to have time and mandate to develop others, and that treats AI fluency development as an ongoing function rather than a project with a defined endpoint. Organizations that build this structure find that their AI capability grows faster than their headcount, which is the signature of compounding. Organizations that rely on external hiring find that their AI capability grows no faster than their hiring, which has a ceiling set by the size and quality of the external market.