Model Context Protocol (MCP)

Model Context Protocol (MCP) is an open standard that enables AI assistants, such as ChatGPT, Claude, Google Gemini and Microsoft Copilot, to securely connect to external systems, data sources and applications. Whereas a traditional API connects software to software, MCP is specifically designed to provide AI models with standardised access to up-to-date information and functionality.

Through an MCP server, organisations can make their business data, databases, applications and other software available to AI assistants. This enables AI assistants to retrieve information, search data and perform actions, while the organisation retains full control over which data and functions are accessible and under what conditions.

By communicating directly with external systems through MCP, AI assistants can work with real-time information and functionality without requiring the AI model to be retrained.

Example
Retriever provides an MCP server that enables AI assistants to access Retriever's up-to-date media and advertising database directly. This allows users to ask an AI assistant questions about available advertising opportunities, rates, audience reach, media types and media specifications. The required information is retrieved from the Retriever database in real time, ensuring that users always receive the most up-to-date information.