ICE opens fixed-income pricing to major AI platforms

Intercontinental Exchange announced licensed users can access its fixed-income pricing data and valuation methodologies on leading AI platforms through the Model Context Protocol, with access currently enabled in Anthropic’s Claude. The connection lets AI systems query ICE datasets while preserving the company’s data hierarchies, valuation methods and user permissioning.

The Model Context Protocol (MCP) is an open standard developed by Anthropic and governed by the Agentic AI Foundation. MCP links large language models and AI applications to enterprise data systems and supports multiple AI services, including Claude, ChatGPT, Gemini and Microsoft Copilot. ICE said the MCP deployment creates a protected knowledge layer that prevents exposure of raw valuation methods while allowing AI platforms to use ICE data to inform responses.

ICE is offering daily end-of-day fixed-income evaluations covering more than 3 million instruments across over 150 countries and more than 80 currencies. An enhanced transparency feed provides roughly 160 data points per identifier, including reference data, price movement commentary, mortgage-backed securities assumptions and trade or quote information. Standalone licensed products available through the protocol include an ICE AAA municipal bond curve and U.S. Treasury benchmark data. The feed also incorporates bond trade data reported to FINRA TRACE for corporate, agency and securitized products and to MSRB RTRS for municipal trades.

According to Chris Edmonds, president of ICE’s Fixed Income and Data Services, the integration enables customers to bring ICE data and analytics into their AI workflows while using MCP’s permission controls to bill for the value delivered rather than the compute consumed. Edmonds added the approach is intended to provide a more predictable and transparent way to incorporate ICE data into decision-making workflows.

ICE is also introducing AI-based products in fixed income. One example is ICE Compass, an analytics platform that ranks trading counterparties and estimates pre-trade prices for buy-side desks. ICE Compass combines a customer’s real-time and historical trading data with ICE market data, pricing streams and daily flows of bids, offers and indications of interest to model intraday movements, trading costs and trading behavior. The tool is designed to help with counterparty selection and pre-trade cost analysis.

ICE framed the MCP integration as a mechanism for licensed users to embed market data into AI-assisted workflows while keeping proprietary valuation methods protected and retaining control over who can access specific data elements. The company said the open MCP standard should enable compatibility with additional AI platforms as they adopt the protocol.