Artificial intelligence has changed how we work. Most people in finance have experienced this first-hand — drafting documents faster, summarising long reports, generating analyses that used to take hours. But there is a limitation that almost nobody talks about. AI, as most people use it today, is isolated.

It lives in a box. It can process the text you give it. It cannot reach your systems, your databases, your deal flow, or your internal records — unless someone builds a specific, custom connection between the AI and each of those systems. And that is expensive, fragile, and does not scale.

The Problem Has a Name

Before November 2024, if a firm wanted to connect an AI assistant to, say, its CRM, its portfolio management system, and its compliance database, it needed to build three separate integrations — one for each tool. If the firm then wanted to use a different AI model, it would need to rebuild all three. This is what Anthropic called the N×M integration problem: N tools multiplied by M AI systems equals an unmanageable number of custom connections.

In November 2024, Anthropic introduced the Model Context Protocol — MCP. It is an open standard that provides a universal interface between AI systems and external tools. Instead of building custom connectors for every combination, each tool builds one MCP server, and every AI system builds one MCP client. The equation collapses from N×M to N+M. More importantly, the connections become standardised, reliable, and reusable.

What MCP Actually Does

Think of MCP as a universal translator. An AI model, at its core, can only read and write text. When it needs to interact with an external system — a database, an API, a messaging platform — it needs to know what that system can do, what information to provide, and how to interpret the response. MCP gives it that instruction set.

An MCP server sits in front of a tool or data source. It gives the AI model a structured description of what is available: what actions it can take, what parameters it needs to provide, and what it will get back. The AI does not need to understand the API directly. The MCP server handles the translation. This is why MCP has been likened to a USB-C port for AI — a single standardised connection point that works across an entire ecosystem.

Why Finance Is Behind

Since its launch, MCP has been adopted by OpenAI, Google DeepMind, and Microsoft. As of early 2026, there are over 10,000 active public MCP servers and tens of millions of monthly SDK downloads. The protocol is now governed by the Agentic AI Foundation under the Linux Foundation — an indication of how seriously the broader technology industry is taking it.

And yet, the financial services industry has been slow. Most firms are still at the stage of testing AI for internal productivity — writing assistance, document summaries, basic search. Very few have considered what it means to let AI interact with their actual operational infrastructure. Fewer still have built the protocol layer that makes that safe, standardised, and scalable.

This Is an Infrastructure Question, Not a Technology Question

The firms that are thinking about MCP today are not asking whether AI is useful. That question is settled. They are asking a harder question: how do we build the infrastructure that lets AI reach our data, our systems, and our partners — reliably and securely?

That question has real implications for private markets, where the flow of information between counterparties is fragmented, manual, and largely invisible. We have been thinking about that problem for a while. In the next post in this series, we go into what the infrastructure layer actually looks like and why it matters specifically for how private market information moves.

--

DISCLAIMER

The information provided is for informational and analytical purposes only and should not be construed as financial advice or an offer or an invitation to buy or sell any securities or related financial instruments. It does not constitute a solicitation, recommendation, or endorsement of any particular security, investment product, or strategy. Investors should seek professional advice of their own before making any investment decisions. The views expressed do not necessarily reflect those of EVIDENT or its affiliates. Readers should independently verify any claims and seek appropriate professional advice before making decisions. For Professional Investors only, as defined under Cap.571 Securities and Futures Ordinance in the laws of Hong Kong Special Administrative Region of China.‍