Function Launches Health Data Connector for ChatGPT, Claude and Perplexity

The connector is designed to give generative AI models access to individual health data rather than relying primarily on generalized population averages.

Function has launched a secure health data connector that allows members to bring longitudinal laboratory results and clinician-reviewed notes into third-party AI platforms, including ChatGPT, Claude and Perplexity. The connector is designed to give generative AI models access to individual health data rather than relying primarily on generalized population averages.

The connector provides structured information from more than 160 diagnostic laboratory markers covering cardiovascular health, metabolism, hormones, thyroid function, nutrients, heavy metals, and cancer screening markers. It also incorporates clinician-reviewed notes to provide additional context around individual results.

According to Function, more than 80% of its members have at least one biomarker outside its optimal reference range. Examples include elevated ApoB, subclinical thyroid dysfunction, and iron deficiency, which can require individual context when interpreting health information.

The new system replaces manual entry of laboratory results and static PDF uploads with a governed data-sharing pipeline. Members can authenticate and authorize access to their health information across supported AI platforms and query longitudinal trends.

For Claude, Function supports connectivity through remote Model Context Protocol (MCP) connectors. The company also cites integrations with ChatGPT Health and apps, as well as Perplexity, enabling users to choose among multiple AI platforms.

The connector builds on Function’s Private AI Chat and its Medical Intelligence Lab (MI Lab) architecture. It is designed to remain model-agnostic, allowing users to switch between foundation models instead of keeping their health data within a single AI ecosystem.

Function said users retain ownership of their health records and can revoke permissions for data access. The system is intended to give AI models access to both numerical laboratory values and clinician-reviewed annotations when responding to health-related queries.

“AI can empower us to understand and take action in so many aspects of our lives. Yet when it comes to health, AI has been missing the context that matters most: our biology,” Function said in announcing the connector.

Stay tuned for more such updates on Digital Health News