When we launched the enterprise AI program, the technology wasn't our constraint. The models existed. The infrastructure was being built. The data was available. The constraint, as it almost always is, was organizational readiness. How do you take 88,000 employees across R&D, commercial, manufacturing, and enabling functions and build the AI literacy, the cultural permission, and the practical capability to actually use AI at scale?
The answer we built was an AI Academy. By the time we had realized $1.5B in recurring annual value from our AI portfolio, we had achieved more than 3,000 course completions, certified 350+ AI Champions embedded across every major business unit, and formalized a certificate program in partnership with Stevens Institute of Technology that enrolled 500+ employees. This is what we learned.
"Culture eats AI strategy for breakfast. You can have the best models, the best data, and the best governance framework — and still fail if the organization doesn't know how to use it, trust it, or improve it."
Why Most AI Training Programs Fail
Most corporate AI training programs make one of two mistakes. The first is being too technical: building content for data scientists and engineers while leaving business users without the vocabulary or context to apply AI to their actual work. The second is being too superficial: producing awareness content that generates enthusiasm without building the practical capability to do anything with it.
Both mistakes share a common root: they treat AI readiness as an information problem rather than a behavioral change problem. You can inform people about AI in a one-hour webinar. Building an AI-ready organization requires something deeper: changing how people think about their work, what problems they bring to technology, and how they evaluate the outputs they receive.
The Architecture of Our Academy
We designed the AI Academy around three distinct learner profiles, each with a different role in the AI value chain:
AI Consumers: The Broad Foundation
This was the largest group: employees across every function who would use AI-powered tools in their daily work, whether or not they knew it. For this group, the curriculum focused on foundational AI literacy: what AI is and isn't, how to evaluate AI outputs critically, how to identify use cases in their own work, and how to raise concerns about AI decisions that affected them. The goal was not to turn everyone into an AI practitioner but to build a workforce that could work effectively alongside AI systems.
AI Champions: The Embedded Network
The AI Champions program was the structural innovation that made the Academy work at scale. Rather than trying to push AI capability from the center outward, we identified and developed high-potential individuals within each business unit who could serve as local leaders and amplifiers. Champions went deeper on the curriculum, covering data governance, prompt engineering, use case development, and change management, and were given visibility into the broader AI portfolio so they could connect local opportunities to enterprise capabilities.
The Champions network became a distributed intelligence system. When a commercial team in Europe had a question about using AI for forecasting, they didn't need to find the central AI team. They had a local Champion who could assess the question, connect them to relevant tools, and escalate if needed. When the central team needed feedback on a new tool's adoption, Champions were their first call.
AI Practitioners: The Deep Technical Layer
The practitioner curriculum was designed for data scientists, engineers, and product managers who were building and deploying AI solutions. This included technical training on model development, deployment, and monitoring, as well as deeper content on AI governance, responsible AI principles, and the business case development process required to move solutions through our stage-gate framework.
The most important architectural decision we made was building the Champion network before we needed it. Change management infrastructure must be built ahead of the change it's meant to support. By the time major AI deployments were rolling out, the Champions were already embedded, trusted, and capable.
The Stevens Institute Partnership
One of the most meaningful investments we made was partnering with Stevens Institute of Technology to offer a formal AI Certificate program for employees. This wasn't a marketing exercise. It was a deliberate signal that AI capability was a genuine career differentiator within the organization, and that the company was willing to invest in developing it to a credentialed level.
More than 500 employees enrolled. The program combined academic rigor with practical application, and the cohort model created learning communities that outlasted the curriculum itself. Alumni of the program became some of the most effective AI Champions in the organization, because they had both the technical depth and the organizational relationships to drive adoption.
What Worked, and What We Had to Learn
The most important lesson from building the Academy was about sequencing. We initially underweighted the consumer layer, assuming that AI tools would be self-explanatory enough that broad literacy wasn't necessary. We were wrong. The first wave of AI deployments generated significant skepticism in business units where employees didn't understand what the tools were doing or why they should trust the outputs. We accelerated the consumer curriculum investment and saw adoption rates improve meaningfully.
The second lesson was about measurement. We tracked course completions and Champions certified, but we were slower to connect Academy activity to business outcomes. When we built tighter linkages and tracked which business units with high Champion density had faster AI adoption rates and higher value realization, the case for Academy investment became much easier to make.
Building an AI-ready organization is not a one-time project. It is an ongoing capability that must evolve as the technology evolves, as the organization's AI portfolio matures, and as the competitive landscape shifts. The Academy we built was a starting point, not a destination. But it was the infrastructure that made everything else possible.