Bloomberg launches Enterprise MCP for AI data access

Bloomberg has launched Enterprise MCP for its Data Licence Plus service, designed to connect Bloomberg data with clients' artificial intelligence applications.

The system gives AI agents access to licensed Bloomberg data through a standardised interface covering more than 100 million securities and over 50,000 fields.

The offering is intended to address a problem that has emerged as financial firms test generative AI tools: models may be able to reason over information, but often lack enough context to interpret market and reference data correctly.

Enterprise MCP is an access layer over Data Licence Plus, Bloomberg's newer Data Licence product. It combines metadata on data fields with semantic search tools and workflow-specific functions, allowing users to request information in natural language instead of relying on field codes or manual identifier matching.

The service can provide details including what a data field represents, how it was calculated, when it applies, and qualifiers such as currency, exchange, price type, period and date. It also includes entity resolution tools designed to map names, tickers or descriptions to the correct instrument and its links to the issuing company.

Bloomberg is offering the product as an enterprise service in which it hosts and manages the access layer, while clients retain control over the AI agents and applications connected to it, including the models, prompts and instructions used in their workflows.

That structure reflects a broader approach emerging in financial services, where firms want to use large language models and agent-based systems but remain cautious about data governance, entitlements and audit requirements.

Under the new service, Bloomberg validates a firm's entitlements before returning data in response to an agent request under standard Data Licence rights. Requests are resolved against Bloomberg's Operational Datastore, while tool-level limits cap the number of securities and fields returned in a single call.

Workflow focus

The initial release includes what Bloomberg calls workflow skills, or reusable procedures tied to common financial tasks. These cover point-in-time universe retrieval, corporate action adjustment and revision history.

The first set of skills will also help users identify outliers in a group of tickers, review trades that breach accepted price deviation thresholds, and assess securities and trades for possible sanctions exposure.

Bloomberg is presenting the service as a way to reduce manual steps in research, portfolio management, risk and operations. In practice, that means reducing the work involved in locating the right dataset, matching identifiers and checking whether a figure is fit for a given task.

"As generative AI adoption in financial services has grown, the bottleneck facing financial institutions has shifted from model capability to data readiness," said Tony McManus, Global Head of Enterprise Data and Indices at Bloomberg.

"We've seen agents that reason well but still can't answer a basic question about a position, because the data reaching them carries no indication of what it actually represents. On its own, a last price doesn't say what currency it's in or what kind of price it is. Connections were never the hard part. Readiness is, and that's what we built Enterprise MCP to solve," McManus said.

Single interface

The service provides a single interface across pricing and reference data, company fundamentals and structure, economics, and alternative data available through Data Licence Plus Live and the Qube Datastore.

Its tool design also avoids creating separate tools for each asset class. Instead, a single security discovery tool and a single field discovery tool are intended to cover the broader content set.

The launch comes as data providers, banks and financial software firms race to position themselves in the growing market for AI tools in regulated industries. In finance, one of the central challenges has been less about access to large language models and more about making underlying market data understandable, attributable and usable within controlled workflows.

Bloomberg said its engineers have been involved in proposals related to governance controls within the Model Context Protocol and in methods that allow servers to present tailored toolsets to different clients. It added that it leads the MCP Financial Services Interest Group and holds seats on the board and Technical Committee of the Agentic AI Foundation, which oversees the protocol as an open standard.

Clients can license specific content domains under standard Bloomberg Data Licence rights.