Every week, I talk to executives who are grappling with a new frontier in technology spend: the AI token model. As platforms like ServiceNow and Microsoft Copilot shift to this model, there's an emerging pattern of AI token sprawl, reminiscent of the cloud sprawl we witnessed with AWS and Azure more than a decade ago. Companies that don't get ahead of this shift will be left scrambling to regain control over their technology budgets.
The problem isn't merely the transition to a token economy. It's the lack of visibility and control over how these tokens are accumulated and used across departments. Just as cloud costs spiraled unexpectedly, AI tokens can become a hidden drain on resources if not managed strategically. This isn't about a single platform; it's about an entire ecosystem of AI solutions, from Anthropic to OpenAI, positioning themselves as foundational technologies for enterprises. Each one is tempting companies to build enterprise value on their platforms, subtly increasing dependency and, subsequently, costs.
The root cause lies in the strategic playbook these AI companies are borrowing from the cloud giants. AWS famously operated as a loss leader, enticing companies to build on its platform before gradually increasing prices to turn a profit. Today, AI providers are doing the same. They offer attractive initial pricing to encourage adoption, but as dependency grows, so does the risk of price hikes, especially if these companies go private and investors demand returns.
So how do you navigate this landscape effectively? First, establish a governance framework that provides visibility into token usage across your organization. Know which departments are using which platforms and for what purposes. Second, apply the principles of FinOps to AI—forecast your token usage, optimize consumption, and continuously monitor for inefficiencies. Third, negotiate with vendors for favorable terms before you become locked into their ecosystems. Finally, ensure that you have a solid ROI model for any AI investment, justifying the spend with measurable outcomes against your strategic objectives.
The counterintuitive move here is to not just focus on immediate cost savings but to think long-term about the strategic value these AI platforms bring. It's tempting to cut corners now to reduce spend, but doing so without a strategic framework could cost more in the long run. The real opportunity lies in leveraging these technologies to drive innovation and competitive advantage, not just cost control.
As you navigate this evolving landscape, remember that AI token economics is more than just a cost issue; it's a strategic one. By reframing your approach from a reactive cost-cutting measure to a proactive value-driven strategy, you can turn what seems like a challenge into a competitive advantage. This isn't just about surviving the AI token wave—it's about riding it to new heights of innovation and growth.