Unleashing Enterprise AI Value in Finance: The CFO’s Guide

It’s quite possible that the most important AI development of 2025 wasn’t a new GenAI model… It was a protocol.

The next wave of AI value in finance won't come from bigger, smarter models. It will come from giving those models a common way to understand context, invoke tools, and act inside enterprise systems.

In November 2024, Anthropic released the Model Context Protocol (MCP) as an open standard.¹ By March, OpenAI had announced support. Google DeepMind followed in April. At its Build conference in May, Microsoft committed to migrating all Dynamics 365 agents to MCP by year’s end.² And in August, NetSuite launched its AI Connector Service with native MCP support.³

That’s a remarkable adoption curve for infrastructure that most business leaders have never heard of.

What Is It?
Think of MCP as the USB-C for AI.²¹ It’s a universal connector that lets any compatible AI assistant access any compatible data source through a single, standardized interface. Before MCP, connecting an AI tool to your ERP meant custom integration work for every combination of assistant and system. Expensive. Fragile. Ungovernable.

MCP changes that equation. AI assistants can be powerful tools, but until now they’ve been largely disconnected from enterprise data. The workarounds we’ve all used (copy-paste, manual exports, asking someone to pull a report) create friction and risk. MCP creates a governed bridge.

This is important for CFOs because it finally answers the question that’s stalled AI adoption in finance: How do we give AI access to our data without losing control?