One of the biggest challenges companies still face in scaling AI and generative AI (GenAI) today? People. Specifically, finding talent with the right combination of AI expertise, business acumen, and the ability to work effectively with AI agents — so they can solve real business problems with speed and at scale.
Some companies try to shortcut the process by hiring so-called “unicorns” — professionals who are equally fluent in machine learning models and strategic decision-making. But these profiles remain incredibly rare and expensive, especially as demand for AI talent continues to outpace supply. While the job market has evolved since the GenAI wave, organizations are still competing for a limited pool of experienced data professionals — and it's clear that hiring alone won’t solve the talent gap. As McKinsey has emphasized in recent years, you can’t hire or outsource your way out of your tech talent problem.
However, there is a proven alternative that we’ve seen succeed in many companies and industries: Upskill your current talent, monitor their performance, and nurture your AI rising stars. The not-for-profit R&D firm MITRE and a Georgetown University think tank recommend that approach for the U.S. military and estimate that 20% of the military civilian workforce — 157,000 knowledge workers — are candidates for AI upskilling. They wrote, “It takes an entire team –– from the coders to the end users –– to achieve success. Therefore, we interpret an AI workforce to include AI technical and non-technical talent.”
Develop Unicorn Teams Not Unicorn People
The approach has worked in the private sector across many industries. Eighty-five percent of companies that have successfully scaled AI use interdisciplinary development teams. Companies whose transformations succeed are nine times more likely than others to engage frontline workers in the transformation. PepsiCo says to stop hiring unicorns because “what you need is cross-functional AI teams.” Develop unicorn teams not unicorn people. We’ve seen it work firsthand with many organizations that use Dataiku, The Universal AI Platform™:
- “Dataiku frees us from seeking ever-elusive ‘unicorns’.” Sarah Cullem, Clorox Director of Direct to Consumer Analytics & Data Science
- Dataiku is “a central environment that very technical and non-technical people can collaborate in. We don’t throw things over the wall anymore.” Chris Kakkanatt, Pfizer Senior Director and Team Leader of Data Science
- “It's all about the wisdom of many people.” Craig Turrell, Standard Chartered Bank Head of Plan to Perform Data Strategy & Delivery
- A North American bank upskilled hundreds of workers. The first year 20% of their projects used ML but that leaped to 70% the second year.
- A consumer packaged goods company upskilled 250 people (85% of whom were non-coders, only 4% were data scientists) and generated $700 million in value the first two years.
- The semiconductor manufacturer NXP upskilled 200 employees, drastically reduced the time to detect manufacturing defects, and saved millions of dollars.
Sounds great, right? Send everyone some YouTube and Coursera links, show them where to download open source Python tools, and you’re on your way to AI value. Change is more difficult than that. Most companies know they have tech talent gaps, including in AI, but 87% of executives say they are not prepared to close it. This gap is compounded by AI being one in a long series of changes:
Such change is difficult both for individuals and organizations, so much so that we hear managers complain of “transformation fatigue.” The goal of this article — first in a series — is to increase your odds of success by presenting best practices for upskilling a large portion of your data analytics workforce. The next article, 5 AI Operating Models That Enable Scalable Success, cover five organizational structures that companies employ as they mature AI/ML and three initiatives we recommend to drive adoption and ROI.
The structure that is best for you depends on your current AI maturity, your ROI goals, risk tolerance, and timeframe.
Upskilling Opens the Door — Structure Unlocks the Value
The AI talent gap isn’t just a hiring problem — it’s a structural challenge that demands a new mindset. The companies leading in AI and GenAI aren’t chasing unicorns; they’re developing unicorn teams by empowering existing employees.
From global banks to consumer goods companies, the evidence is clear: When you invest in your people, value follows. But while upskilling opens the door, scaling it sustainably requires the right organizational support — one that reduces transformation fatigue and enables people to thrive in an AI-first culture.