The Scarcity Of AI Talent: A Problem Of Our Own Making Is A Problem We Can Solve

Let's try something different to address the diversity in AI challenge.

Credit: marrio31

Last year alone the tech sector saw incredible levels of innovation, from digital contact tracing and self-fertilizing crops to harvesting energy from wireless signals. Innovation is what we do. If we preferred the status quo, we likely wouldn’t be leaders in this field. And yet, when it comes to recruiting diverse tech talent — especially in highly specialized areas like AI — we are far less innovative. In fact, I’d argue that when it comes to how we think about our AI talent pipeline, we are stuck in a rut of our own design, one that has left most companies competing for what appears to be a shortage of resources.

But when you really think about it, it’s not as if people with the innate skills to perform AI jobs are a finite resource. The scarcity of AI talent is not an inevitable state of nature — it’s just easy to tell ourselves the story that it is, even as our own actions perpetuate the problem. We act as if the small handful of colleges and universities that have top ranked capabilities in AI education and training are the only places to look for our future AI workforce But as Ruthe Farmer and Dwana Franklin eloquently wrote in TechCrunch late last year, it’s time to stop equating privilege with potential. If we just widen the aperture, we see hundreds of other schools in the country with solid computer science programs already in place. Why not think about how we might bring more AI training to these less privileged but high potential pools of students?

I realize that building undergraduate AI educational programs may seem like academia’s problem to solve, and to some extent, it is. But I caution against this simplistic and convenient stance. In my decades of experience leading technical organizations in corporate America, and more recently, as a social entrepreneur, I’ve seen that there are incredible untapped opportunities for innovation at the intersection of industry and academia — opportunities for partners from each sector to “play” in each other’s backyards, both to enhance students’ education and to develop future employees. Solving the AI talent shortage is just such an opportunity.

Here’s why industry must partner with academia: research shows that independent of a variety of demographic and institutional factors, and independent of technical course work, students are far more likely to be hired (as interns or as employees) if they can show a portfolio of real world work samples — things they’ve built, problems they’ve solved. But here’s the Catch 22: these opportunities require the active participation of industry partners and collaboration with university programs. If our AI recruitment efforts are focused on only a limited number of universities, how are students who do not attend these universities — and who lack relevant personal or professional networks — supposed to gain access to coveted, portfolio-building experiences? In other words, how can a student demonstrate their potential, if they lack the access that comes with privilege?

The good news is that the problems we create, even unintentionally, are problems we have the power to solve. Each one of our companies can resolve at least part of the issue that is fueling the shortage of AI talent; here’s how:

1. Partner to open up new pathways: I know it might sound daunting to completely rethink your recruitment practices. The good news is, there are organizations dedicated to bridging the gap between companies like yours and networks of diverse tech students who attend schools outside of your standard university partnerships. Partner with one of these companies and benefit from their bridge-building expertise.
2. Dedicate resources: There’s no way around it. We must invest in developing this talent pipeline; doing so will pay dividends for years into the future. Working with one of the aforementioned partner organizations, establish an AI recruitment program that prioritizes collaboration with diverse undergraduate students at schools beyond those you’ve traditionally targeted. More specifically, identify real-world problems that teams of these students can use AI technology to help solve, then choose current technical staff members to advise your student teams. In addition to helping to stoke your talent pipeline, you’ll be providing a great leadership development experience for your top ranked up-and-comers.

A number of companies have taken a different approach to solving the AI talent shortage: creating their own AI training programs so that young people bypass college altogether. While this is indeed a valid solution for some, and certainly removes the barrier of the high cost of college tuition, I caution against putting all of our eggs in that basket. For one thing, we have tens of thousands of undergraduate computer science students in this country; let’s get creative about how to match this supply of talent-in-waiting to meet our demand. What’s more, while corporate training programs provide the technical skills necessary to perform certain jobs, I wonder how well they equip young people for long-term success, particularly when it comes to ascending to leadership roles (and salaries). Either way, as innovators, I think we can agree that having more potential pathways to a solution is a stronger strategy than betting on one approach alone.

If we are willing to leverage our signature skill at thinking outside the status quo, we have the power to create the supply to meet the exploding demand for AI talent. But like any other business problem we try to solve, it will require the allocation of resources and a commitment to innovative solutions.

Follow me on Twitter or LinkedIn. Check out my website.