The evolving cost strategy for enterprise AI scaling

Organizations are shifting their focus from evaluating AI's effectiveness to managing its costs as they scale AI across large teams. Jason Banta, vice president of global go-to-market compute sales at Qualcomm Technologies, highlights that agentic AI workflows can require up to 20 model interactions per task, complicating budgeting. Traditional fixed-seat cost models are no longer sufficient, as AI usage now relies on token-based billing. Gartner notes that agentic models consume 5 to 30 times more tokens than standard chatbots. Despite falling token prices, higher consumption is expected to increase overall costs. Companies are adopting tokenomics strategies to optimize AI budgets, balancing cost, performance, and security. On-device AI is gaining traction as a way to reduce costs, energy use, and carbon footprint, with Qualcomm emphasizing its role in enabling efficient AI deployment.