Everyone wants an AI strategy. But according to business leader Pankaj Prasoon, that’s like asking for a “spreadsheet strategy” — it completely misses the point. “Enterprise AI is a practice, it’s not a theory,” says Prasoon, Senior Director of Finance and AI at Microsoft. “I’ve guided organizations through the realities of applying AI into finance, ERP, and supply chain where the cost of failure is measured in billions. AI isn’t the strategy. It’s the amplifier.”
Prasoon has led global transformation programs at scale, including multibillion-dollar cloud migrations and the integration of AI into mission-critical operations. Having written extensively on enterprise AI, including his latest book Strategic AI in ERP, he stresses that the most valuable conversations are not about which model to adopt but about how businesses can translate AI’s promise into measurable impact.
Don’t Fall for the AI “Quick Fix”
Prasoon cautions against treating AI as a magic bullet. Too often, companies spin up tactical projects masquerading as strategy. In those cases, AI doesn’t fix the flaws. It only magnifies them. The first pitfall is poor data foundations. “Garbage in, garbage out, that too at scale,” he explains. Imagine a retailer trying to predict sales with bad data: the system ends up recommending winter coats in July. Another trap is what Prasoon calls the “pilot to production” problem. A small innovation team dazzles the board with a demo, but when the shiny new model has to connect with a 20-year-old accounting system, the project stalls for months and eventually gets shelved. Equally important is culture. “AI is a team sport,” Prasoon emphasizes. “Quick fixes assume technology alone will solve everything, but organizations often fail to prepare their people. Without training, communication, and a clear purpose, employees fear AI as a cost-cutting tool. That resistance can quietly kill adoption.”
Building on a Strategic Foundation
Prasoon’s advice: start with the business problem, not the tool. “Before you consider AI, define the core challenge. Is it customer churn? Supply chain efficiency? Unclear market signals? Too often, companies deploy AI where a process redesign would suffice, just to check the AI box.” He identifies four pillars that underpin successful adoption: process discipline, high-quality data, modern infrastructure, and governance. Without these, organizations will stumble.
Consider a logistics company:
That’s the amplifier effect in action. “The goal isn’t to replace jobs,” Prasoon says. “It’s to use AI to free people to do more meaningful work.”
From Initiative to Infrastructure
Looking ahead, Prasoon predicts a shift from experimentation to embeddedness. “In the next half decade, AI will stop being treated as an initiative and will start becoming infrastructure. It won’t be a sidecar strategy but an amplifier woven into every decision, every product, and every customer interaction.”
This evolution will mean AI is less about chatbots and productivity hacks and more about becoming a trusted system of intelligence at the core of finance, HR, and supply chain. Organizations that succeed will be those that use AI to build proprietary data moats and competitive advantage. “The future isn’t man versus machine. It’s man with machine,” Prasoon explains. “Tomorrow’s financial analyst won’t spend their day building pivot tables. They’ll use AI to crunch the numbers and focus on asking the right questions: ‘What’s the root cause of this anomaly?’ or ‘What happens if we shift our pricing strategy?’ Their value will come from critical thinking, strategic insight, and emotional intelligence — uniquely human skills.”
Innovation Above the Infrastructure
For Prasoon, the true opportunity lies not in the foundation models themselves but in what gets built on top of them. “Refrigerator companies don’t make the most money from refrigeration. Coca-Cola does, because its entire product depends on it. You’ll never drink a warm soda. Similarly, Microsoft, Google, and Amazon will make the refrigerators of AI — the GPTs and Geminis. But the game changers will be the companies that build something entirely new on top.”
He points to industries already moving in this direction: drug discovery firms using AI to slash R&D timelines, creative startups inventing new genres of art, logistics players reimagining global supply chains. “The real value won’t be captured by those building large language models but by those who use them to create new business models and products. The companies that win in the age of AI won’t be the ones with the best models — they’ll be the ones that have built the best businesses on top of them.”
For more insights from Pankaj Prasoon, connect with him onLinkedIn or visit his website.
