The Pulse
Waters Ships an MCP Interface for Laboratory Software
Waters says its AI Link software provides a conversational route to selected waters_connect laboratory tasks using the Model Context Protocol. The initial tool set focuses on quantitative acquisition and processing, with actions governed by
AI.info Team · Written with AI ·
Waters says its AI Link software provides a conversational route to selected laboratory tasks in waters_connect through the Model Context Protocol (MCP). The company describes AI Link as an MCP server that converts platform capabilities into defined tools, rather than granting an AI model unrestricted access to the software.
Waters published the announcement on October 13, 2026. It presents the interface as an additional way to use existing laboratory capabilities, not a replacement for scientific applications, validated assets or domain expertise. The company’s stated aim is to make routine steps easier to reach while keeping the underlying work within the platform.
AI Link targets quantitative workflows
AI Link currently brings together capabilities for quantitative acquisition and processing. Waters lists tasks including creating and editing sample lists, working with acquisition methods, assigning instruments, accessing analytical results and generating reports.
The conversational interface is aimed at tasks that can require users to move among applications, menus and setup steps. An analyst can describe an intended task in natural language, and the platform’s defined tools translate that request into actions. Waters says teams can reuse existing methods and sample structures, building on validated work instead of starting again.
The company describes potential benefits for different users. Analysts may have a simpler route through routine workflows; scientists and method developers may spend less time on repetitive setup; and laboratory managers may see more consistent execution and less reliance on a small number of experienced users. Waters also points to the learning curve for newer analysts as one reason routine tasks can take time and require help from colleagues.
Access stays within platform permissions
Waters emphasizes limits on what the interface can do. Each tool is explicitly defined and constrained, and actions remain subject to the platform’s authentication, permissions and access controls. Users can perform through AI only actions they are already authorized to perform in waters_connect.
“Every action remains subject to the platform’s authentication, permissions, and access controls.”
Dave Leitham, Waters blog author
Waters says this governed approach distinguishes AI Link from an uncontrolled agent. Natural-language requests provide another interface to platform capabilities, not a shortcut around product ownership, validation or laboratory policy. The domain applications remain the authoritative environment for detailed scientific work.
MCP provides an extensible foundation
AI Link is built on MCP, an open standard that enables AI assistants to connect with software capabilities in a consistent way. Waters says MCP-compatible assistants can access published tools without a separate custom integration for every use case. The announcement does not identify specific compatible assistants.
The company describes a plugin model through which additional waters_connect capabilities could become available using the same governed mechanism. Its longer-term goal is broader workflow coverage, from method development through acquisition and processing to results retrieval. Waters does not provide a rollout schedule for those future additions in the announcement.
For now, the announced scope is quantitative acquisition and processing, with a defined set of tasks operating within the permissions already assigned in waters_connect. Waters frames the product as a practical conversational entrance to existing laboratory workflows, while leaving scientific applications and validated assets in place.