Why AI agents Cost So Much

AI agents can quickly deliver business value, but as usage scales, many organizations are discovering that intelligent experiences come with real operational costs. In this episode of the Azure Essentials Show, host Takuto Higuchi sits down with Sindhura Bandhakavi, Director of Engineering, Microsoft CoreAI, to explore the emerging discipline of AI token economics, or "tokenomics," and what it takes to optimize agent performance without sacrificing quality. Through practical demonstrations of capabilities such as Model Router and Agent Optimizer in Microsoft Foundry, they unpack the factors that drive AI costs, explain why reducing tokens alone is not always the right strategy, and show how organizations can treat AI as a managed investment. Whether you're building your first AI agent or looking to improve ROI across a growing portfolio of agents, this episode provides actionable guidance for balancing cost, performance, and scalability. Three things you will learn - Why AI agents can become expensive to operate at scale and the key factors that drive token consumption, model costs, and overall AI ROI. You'll learn why simply reducing tokens is not always the best optimization strategy. - How to use Azure AI Foundry capabilities such as Model Router and Agent Optimizer to improve agent efficiency, automatically select the right model for each request, and continuously enhance agent performance over time. - How to use Azure AI Foundry capabilities such as Model Router and Agent Optimizer to improve agent efficiency, automatically select the right model for each request, and continuously enhance agent performance over time. - How to manage AI as a business investment by measuring costs, evaluating agent effectiveness, establishing controls for usage spikes, and ensuring every AI request delivers the right balance of quality, performance, and cost.   Resources  - The economics of agent optimization: from pilots to measurable returns   https://azure.microsoft.com/blog/the-economics-of-agent-optimization-from-pilots-to-measurable-returns/ - Bridging the AI divide: How Frontier Firms are transforming business  https://blogs.microsoft.com/blog/2025/11/11/bridging-the-ai-divide-how-frontier-firms-are-transforming-business/ - Model router for Microsoft Foundry concepts   https://learn.microsoft.com/azure/foundry/openai/concepts/model-router - How to use model router for Microsoft Foundry  https://learn.microsoft.com/azure/foundry/openai/how-to/model-router - Agent optimizer in Foundry Agent Service overview https://learn.microsoft.com/azure/foundry/agents/concepts/agent-optimizer-overview - Introducing Agent Optimizer in Foundry Agent Service  https://devblogs.microsoft.com/foundry/agent-optimizer-build2026/ - What's New in Microsoft Foundry, June 2026  https://devblogs.microsoft.com/foundry/whats-new-in-microsoft-foundry-june-2026/ - 2025 Stack Overflow Developer Survey https://survey.stackoverflow.co/2025/ai Related episodes  - Cut AI Agent Costs with TokenOps https://www.youtube.com/watch?v=B4yovi_znxE - Getting Started with the Microsoft Agent Pre-Purchase Plan https://www.youtube.com/watch?v=0-RN85uZzTY - Watch more episodes of the Azure Essentials Show https://aka.ms/AzureEssentialsShow Connect  -Takuto Higuchi https://www.linkedin.com/in/takuto-higuchi-626681135/ -Sindhura Bandhakavi https://www.linkedin.com/in/sindhura-bandhakavi/  Chapters  00:00 In this episode 00:26 Welcome to Tokenomics 01:00 Why AI Agents Cost More Than You Think 04:00 The Hidden Cost of Chasing Fewer Tokens 07:00 Measuring ROI for AI Agents 10:00 Choosing the Right Model with Model Router 12:00 Live Demo: Model Router in Action 15:30 Improving Agents Over Time with Agent Optimizer 18:00 Live Demo: Optimizing Prompts and Agent Performance 21:20 Building a Continuous Optimization Flywheel 22:20 Your AI Budget Review Checklist 23:16 Key Takeaways and Resources 22:53 Getting Started