LabKey launches AI tools for governed lab data

LabKey has launched LabKey AI, a plain-language query layer built into its regulated lab software stack in Seattle. The system is designed to return traceable answers from live data while preserving existing permissions, audit trails and validation controls.

Why it matters: - Lab teams can ask questions in plain language and get answers grounded in governed, structured data instead of approximate AI output. - The launch targets regulated scientific work, where permissions, auditability and traceability are essential. - LabKey AI could reduce dependence on specialists for routine reporting and data lookup.

What happened: - LabKey today announced the availability of LabKey AI. - The product is an intelligence layer built into LabKey LIMS, Biologics LIMS, Sample Manager and SDMS. - The system is available now. - LabKey says scientists, lab managers and IT teams can use the tool to query, understand and act on lab data using plain language.

The details: - LabKey AI runs on the LabKey data model, the structured and access-controlled foundation used across life and applied science labs. - The intelligence layer inherits LabKey’s existing permission framework. - Every answer comes from a query or API call against live data. - A user with read-only access to a project gets answers from that project only. - LabKey does not use customer data to train general AI models. - Current features include natural language querying of structured lab data, in-product documentation search and a read-only MCP layer for technical users. - Each result is traceable back to the underlying query and reasoning. - A sample question about freezer inventory returns an exact count from live data, not an approximation. - LabKey says any action that writes, modifies or deletes data will require human confirmation. - LabKey says the rules governing those actions live in LabKey, not in the AI layer. - The capability can be disabled for teams that are not ready to adopt it.

Between the lines: - The launch positions AI as a workflow layer on top of existing controls, not a replacement for them. - That approach matters in QA-heavy and regulated environments, where a new AI system can create risk if it sits outside established governance. - LabKey is also signaling that the product is early-stage, with reporting and data management features planned next.

What's next: - LabKey plans to extend the plain-language interface into reporting and data management. - Future additions will keep the same data, security and permissions model. - Any future write, modify or delete actions will still wait for a person’s confirmation.

The bottom line: - LabKey is betting that regulated labs will adopt AI faster if the system is built directly into governed data rather than layered on top of it.