Bloomberg launches enterprise MCP for AI data access

Bloomberg has launched Bloomberg Enterprise Model Context Protocol for its Data Licence Plus service, giving client AI agents access to licensed Bloomberg data through a standardised interface.

The new access layer covers more than 100 million securities and more than 50,000 fields. It is aimed at firms using AI tools in research, portfolio management, risk and operations, where agents need to identify, interpret and retrieve data without manual dataset searches or identifier matching.

The service combines data access with metadata describing what each field means, how it was calculated and when it applies. This includes qualifiers such as exchange, currency, price type, period and as-of date, which are needed if an AI system is to use market data correctly.

The system also includes semantic search tools that let agents find fields and securities through natural-language queries rather than Bloomberg mnemonics. On the security side, the service can map a name, ticker or description to the relevant instrument and its relationship to the issuing company.

Content available through the interface includes pricing and reference data across asset classes, company fundamentals and structure, and economics and alternative data from Bloomberg's Data Licence Plus Live and Qube Datastore services. Clients can license the content domains they need under standard Data Licence rights.

Workflow tools

Alongside the data interface, Bloomberg is introducing reusable workflow functions, which it calls Skills. The initial release includes tools to identify outliers in a set of tickers, review trades that breach accepted price-deviation thresholds, and assess securities and trades for potential sanctions exposure.

The design uses a single security discovery tool and a single field discovery tool across the content set, rather than separate tools for different asset classes. Requests are checked against a client's entitlements before any data is returned, and tool-level limits restrict the number of securities and fields in a single call.

Bloomberg hosts and manages the service, while clients retain control over the AI agents and applications connected to it, including the models, prompts, instructions and other settings used in those workflows.

Tony McManus, Global Head of Enterprise Data and Indices at Bloomberg, said firms' difficulties with generative AI in finance have shifted from model performance to the quality and context of the underlying data.

"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."

Broader push

The launch adds to Bloomberg's wider effort to build products around Model Context Protocol, an emerging standard for connecting AI systems with external tools and data sources. Its engineers are among the main authors of proposals intended to add enterprise governance controls to the protocol and to let servers present different toolsets to different clients.

Bloomberg also leads the MCP Financial Services Interest Group and holds seats on the board and Technical Committee of the Agentic AI Foundation, which stewards the standard.

Bloomberg has used artificial intelligence techniques in financial data and information workflows since 2009, spanning machine learning, natural language processing, information retrieval, time-series analysis, generative models and agentic AI systems. The new product reflects how financial data providers are adapting established licensing and entitlement structures to AI-driven use cases inside banks, asset managers and other market participants.

For those users, the issue is not only whether an AI system can retrieve a data point, but also whether it can determine if the field is relevant, what the figure represents and whether it can be trusted for a given task. Bloomberg's approach is to package data access with descriptive and operational context so client systems can use the information within governed workflows.

Enterprise MCP is being offered as an enterprise product for Data Licence clients, with access governed under existing Bloomberg data rights.