Freedom Opens Brokerage Infrastructure to AI Agents, Signaling Shift Toward AI-Native Finance

Freedom Opens Brokerage Infrastructure to AI Agents, Signaling Shift Toward AI-Native Finance

MCP integration gives AI tools access to brokerage data and selected functions as Freedom24 moves toward a broader European financial ecosystem

Tradernet, the proprietary technology platform behind Freedom24, a European financial technology company, has opened its brokerage infrastructure to AI applications through an integration based on the Model Context Protocol (MCP), allowing users to connect their brokerage accounts to AI tools including Claude Code, Cursor and Codex.

The development marks an early move toward an evolving model of financial technology: rather than customers navigating an app to access individual financial products, AI agents could increasingly become the interface through which customers interact with financial infrastructure.

Through the MCP integration, users can access selected brokerage capabilities using natural-language instructions, including portfolio information, market quotations and historical data, orders and trade history, reports, and price and event alerts.

For example, a user can ask an AI assistant to show their portfolio, retrieve the current price of a security or review their trading history, without navigating the brokerage interface directly.

The significance extends beyond a new way to access a brokerage account. By exposing selected brokerage capabilities as standardized tools for AI applications, Freedom is preparing financial infrastructure for an environment in which AI agents can interact with financial services alongside traditional applications.

From financial applications to AI-accessible infrastructure

MCP is an open protocol that standardizes how AI applications connect to external tools and data. It does not itself provide brokerage functionality, execute trades or replace financial APIs.

Its significance for financial services is that capabilities traditionally accessed through an application interface or bespoke API integration can also be made available to AI systems through a standardized connection. This creates the foundations for a different way of interacting with financial services, in which an AI interface can sit between the customer and the underlying financial infrastructure.

Rather than requiring customers to navigate separate applications to access individual financial products and services, AI agents could increasingly become a layer through which customers access information and authorized financial capabilities across those services.

“The important development is not simply that an AI can answer a question about a portfolio,” said Vitali Zvonar, Chief Product Officer at Freedom24. “It is that financial capabilities can increasingly become accessible through intelligent interfaces. We see this as an important step in preparing financial infrastructure for a world in which AI agents become part of how customers interact with financial services.”

The development forms part of Freedom’s broader technology strategy. Tradernet, alongside the company’s AI capabilities and Neo Compliance technology, is developed in-house, giving the company control over the infrastructure connecting its investment services, technology and operational environment.

From brokerage toward a broader financial ecosystem

The move comes as Freedom pursues a broader European financial-services strategy. In June 2026, the company applied for a banking license in France, potentially expanding its European offering beyond investment services. The company has also outlined ambitions to develop a broader financial ecosystem around its brokerage business.

The strategy builds on the wider Freedom SuperApp model operated by Freedom Holding Corp. in the Group’s largest market, Kazakhstan, where it brings together financial and lifestyle services including banking, payments, investing, insurance, telecom, e-commerce and more. It also reflects a broader view of the European market outlined by Freedom Holding Corp. CEO Timur Turlov in a recent Les Echos op-ed in which he argued that the next phase of fintech competition will be shaped by how effectively financial services are integrated around customers.

The combination of these developments points to a potential evolution of the SuperApp model. The first generation of digital finance moved traditional services online, while the SuperApp model brought multiple financial products into a single application. The emerging AI-native model could make the application itself less important as the primary interface, with an AI agent able to access multiple services through standardized connections.

AI access does not mean unrestricted autonomy

Connecting AI to financial infrastructure also creates a fundamental difference from many other AI use cases: an AI system can potentially interact with information and actions that have direct financial consequences for a customer.

Tradernet’s current MCP implementation is therefore designed around explicit limitations. Trading actions require customer confirmation. Transfers and withdrawals are not available through MCP. API credentials remain under the customer’s control rather than being stored by the connector. The approach reflects a principle that is likely to become increasingly important as AI moves into regulated financial services: making financial infrastructure accessible to AI does not necessarily mean giving AI unrestricted authority over it.

“The question is not only what AI can do,” said Zvonar. “In regulated finance, the more important question is what AI should be allowed to do, under what permissions and with what level of customer control.”

Building for the age of agentic finance

The Tradernet MCP integration represents an early example of a broader architectural shift that could affect how financial services are designed. Under the traditional model, each financial service has its own interface, while under an AI-native model, those capabilities could increasingly sit behind an intelligent interface accessible via AI agent.

MCP is only one component of such an architecture. It does not replace banking infrastructure, financial APIs or regulatory frameworks. Its role is to provide a standardized mechanism through which AI applications can interact with external capabilities.

For financial institutions, that raises a new strategic question: if AI agents become a major customer interface, should financial infrastructure be designed not only for humans using applications, but also for machines acting on behalf of humans?

Freedom’s MCP integration may be an initial step in that direction.