The promise of AI is immense, but Clyde Seepersad believes the way organizations adopt it will determine whether it becomes a productivity booster—or just another short-term cost-cutting tool.

The productivity paradox of the digital age has a name, and Clyde Seepersad knows exactly what it is. As Senior Vice President and General Manager of Education at the Linux Foundation, Seepersad has been watching the numbers with the keen eye of someone who understands that behind every percentage point lies human potential.

“I was looking at this data recently—from the end of World War Two, for 60 years, productivity in the US grew at about two and a half percent a year, and starting in 2000 it dropped to about 1.4–1.5%,” Seepersad explains. The implication is staggering: “So apparently the internet made us less productive, not more productive.”

This isn’t just an academic observation. Twenty years of one percentage point less productivity growth represents a massive economic gap, and Seepersad believes artificial intelligence might be the key to bridging it. But his vision for AI implementation runs counter to the cost-cutting narratives dominating boardrooms across enterprise America.

Rethinking the AI Playbook

The challenge isn’t technological—it’s foundational. “We had a really good playbook and we knew how it worked, and now it feels like somebody set it on fire,” Seepersad observes. Traditional hiring practices, skill-based recruitment, and career development pathways have been disrupted by a technology that’s moving faster than institutional knowledge can adapt.

Consider the hiring dilemma: organizations are looking for candidates with three years of experience building agentic AI systems, but “nobody has three years of experience building agentic AI, nobody has one year of experience,” Seepersad notes. The entire framework for talent acquisition needs reconstruction from the ground up.

His solution involves shifting from skill-specific hiring to something more akin to an apprenticeship model. Maybe it looks less like specific-skills hiring—oh well, I want to go get a software developer with experience in X language—and more like I want to bring somebody in, expose them to the business, and sort of an apprenticeship model,” he suggests.

This approach acknowledges a fundamental truth about AI implementation: context matters more than code. Seepersad illustrates this with a story about a fast-food company that implemented an AI agent for drive-through orders. The system was technically impressive—it could handle accents, map orders to menus, and play back confirmations. But it failed spectacularly when customers wanted to add items after confirming their order.

“Any human who would work for 20 seconds on a line would have told you people change their order,” Seepersad explains. “That context of the business—the abstraction away from building the basic model—is the easy part. Figuring out what the actual process is that your customers are using and making sure you’re meeting those use cases, that’s the bit where humans add value.”

Beyond Cost Cutting

While many organizations view AI through the lens of workforce reduction, Seepersad advocates for a more strategic perspective. “No company ever cost-cut its way to greatness,” he states emphatically. “You cut the easy stuff, then you cut the slightly less easy stuff, and eventually, in the analogy, you’re cutting muscle and bone.”

The real opportunity lies in enabling capabilities that were previously impossible. Seepersad points to breakthrough applications in drug discovery, protein folding, genomics, and real-time customer insights as examples of AI’s transformative potential. “There is amazing power in terms of growth and innovation and outreach. We can’t let ourselves get so distracted by the potential for some short-term cost-cutting gains that we give up the long-term benefits.”

This long-term thinking extends to talent pipeline management. Using a vivid analogy about his recent plumbing experience—where the technician, in his early 60s, was charging $125 an hour plus service calls—Seepersad warns about the consequences of not investing in entry-level talent. “There are no 20-year-old plumbers. There are no 30-year-old plumbers. Terrible for me as a consumer, because if I want an electrician, if I want a plumber—if you don’t feed the pipeline of talent, eventually you pay the price. Eventually your customers pay the price.”

The Education Laboratory

At the Linux Foundation, Seepersad’s team is experimenting extensively with AI across their educational offerings. They’re using AI to aggregate expertise for exam design, reducing the time commitment required from subject matter experts while maintaining quality. “Instead of trying to find 10 experts, we’d say, hey, we need maybe 20 hours of your time over the next 10 weeks. Half of them drop out because they’re like, I can’t commit. If we can instead go to them and say, Hey, this is the first draft AI generated of what we think the core skills are. Help us edit it. Help us make it better. Now their time commitment drops to five hours.”

The approach extends to content creation and infrastructure management, with AI enabling the production of more educational material in formats that meet learner preferences. “We’re producing more content than we ever have because we can, because we can leverage those tools to do AI-generated video and voice-over so that more of it is in the form factor that the younger audience wants, which is five-minute videos.”

Critically, this increased output hasn’t resulted in workforce reduction. “We still have our whole team in place, right? And so it really has been about accelerating productivity,” Seepersad emphasizes. Human expertise remains central to the value proposition—AI simply amplifies their capabilities.

The path forward requires embracing perpetual learning. “We have left behind the world where you could train to be X and do X for 30 years,” Seepersad observes. “We’re all on this journey now of continuous learning, and it’s uncomfortable, because in the old world, I would master a domain, and then I could leverage that mastery.”

The new reality demands comfort with discomfort. “If you get to a point where you feel comfortable that you have mastery, you should be very worried. The world is changing too fast. You have to constantly challenge yourself to find out what’s out there, what’s new, what don’t I know that I should know.”

For organizations, this means recognizing that their existing workforce represents accumulated business context that can’t be easily replaced. “This idea of getting the value of AI is going to require people who actually know my business,” Seepersad explains. “Maybe I should make sure they know how to write a model. Maybe I should make sure they know how to ingest data into a system. Maybe I should make sure they know how to think about the IP implications of loading this stuff in.”

The investment in upskilling may cost thousands of dollars, “but the payoff is immense,” he concludes. In an era where AI promises to transform productivity, the companies that will thrive are those that view their people not as costs to be cut, but as force multipliers to be empowered.

Continuous Learning as a Career Imperative

For individuals, Seepersad’s advice is clear: the era of training for one role and doing it for decades is over. “By the time I master it, it’s outdated,” he says. Success now depends on continuous learning and embracing the discomfort of constant change. For organizations, that means long-term talent strategies—investing in current employees so they can work alongside AI, not be replaced by it.

At the Linux Foundation, these principles are already in practice. AI is used to create first drafts of exam questions and course outlines, allowing experts to spend more time refining content and less time on tedious work. The result? Faster delivery, more engaging materials, and greater alignment with how learners consume information today, such as short-form video.

Seepersad’s closing message is both a warning and an encouragement: if you ever feel comfortable in your mastery, you should be worried. The world is changing too fast for comfort. The winners in the AI era will be those who continually evolve—individuals and organizations alike.