Summary
The financial implications of artificial intelligence in enterprise operations are becoming increasingly complex as companies transition from pilot projects to full-scale implementations. Kevin O'Mara and Kelly Coutinho explore the economic divide between the initial costs of developing AI capabilities and the ongoing expenses associated with their daily use. Unlike traditional enterprise software, where costs are primarily tied to implementation and licensing, AI introduces variable costs based on the consumption of tokens—small units of information processed by models. This shift necessitates a reevaluation of how organizations budget and manage these expenses, particularly as the demand for AI capabilities grows.
Coutinho emphasizes the need for leaders to focus not only on the costs associated with AI but also on the value it generates. As organizations integrate more AI agents and data, the infrastructure must be upgraded to support these systems, which can lead to increased operational costs. The conversation highlights the importance of understanding the balance between the cost of AI operations and the strategic value they can provide, urging leaders to consider how AI can fundamentally change business processes and improve decision-making.
As the discussion progresses, the challenges of forecasting AI-related expenses become apparent, particularly when these costs are intertwined with partnerships and vendor services. Coutinho suggests that leaders should adopt a mindset that views technology as a set of capabilities rather than mere products to be managed. This shift in perspective is crucial for harnessing the potential of AI as a dynamic economic engine, ultimately driving value creation and competitive advantage in an evolving landscape.
Show Notes
For years, enterprise AI has been framed as a build decision. But as agents move into full-scale operations, the real cost is running them every day – and the meter doesn’t stop running. This week, Chief Research Officer Kevin O’Marah sits down with VP, Intelligence & Advisory Kelly Coutinho to unpack the new economics of enterprise AI. They explore why a paper mill and a McDonald’s promotion need very different kinds of AI and compute, what to meter versus where to commit to dedicated capacity, and why what your business knows, not access to frontier models, is what delivers a lasting competitive edge.
More from The Zero100 Podcast
What’s harder: transforming a century-old carmaker for the digital age, or building one from scratch with no safety net? In this special episode, Zero100 Executive in Residence Jim Rowan – former CEO of Volvo Cars and Dyson – reunites with his former Volvo colleague Javier Varela, now Chief Operations Officer at Rivian, to explore how a software-first foundation is paving the way for AI-defined operations. Javier shares how Rivian uses the vehicle itself as part of the factory system, why early architectural choices helped cut the R2’s bill of materials in half, and how its homegrown “Large Driving Model” and custom silicon position the company to win in autonomy.
For years, leaders have talked about how difficult it is to secure funding beyond standard operations, assuming that breaking through requires a massive balance sheet and a deep tech budget. But Zero100’s latest research reveals a surprising truth: the correlation between investment and digital maturity is virtually nonexistent. This week, Chief Research Officer Kevin O’Marah sits down with VP, Research & Advisory Jenna Fink to explore findings from her upcoming report on the future of the orchestrated factory, discussing how to open up the “black box” of production, leverage the agility of smaller footprints, navigate integration hurdles in brownfield sites, and shift toward network-level decision-making.
With AI rapidly flooding the world with text, design, and code, will the market place a premium on human-made craftsmanship and strategic thought, or will efficiency and utility entirely eclipse the human element? This week, Zero100 Chief Content Officer Matt Davis moderates as Lauren Acoba (VP, Head of Research & Advisory) and Justin Gillebo (Senior Director, Research & Advisory) debate the future value of human work. Tune in as they unpack everything from “AI slop” and the hidden costs of unthoughtful content generation to agentic workflows, accountability, and why supply chain leaders must experiment while building rigorous human judgment.
We’ve spent the last decade building better control towers and gathering more data than ever. So why is the supply chain still slowing down when a disruption hits? VP, Research & Advisory Jenna Fink and Senior Director, Research & Advisory Justin Gillebo unpack why visibility isn’t the problem anymore, the tough realities of 3PL data ownership, why physical freight automation (like driverless trucks) might actually outpace our control rooms, and how leaders can finally bridge the gap between control room insights and real-world execution.
This week, Chief Content Officer Matt Davis sits down with Arun Krishnan – SVP, Global Supply Chain & Strategy at AstraZeneca – to explore the Holy Grail of modern operations: the self-healing supply chain. Pulling back the curtain, Arun shares how AI and human judgment are combining to automate the routine, eliminate the guesswork, and protect patient safety at scale, all while bringing a genuine sense of joy back to the team by freeing them up for higher-level work.