What You Should Know
- Function launched a secure health data connector enabling members to bring their longitudinal lab results and clinician-reviewed notes directly into third-party AI platforms, including ChatGPT, Claude, and Perplexity.
- The connector addresses a primary limitation of consumer health AI by replacing generalized population averages with personal biological context, covering 160+ diagnostic lab tests across heart health, hormones, thyroid, metabolism, heavy metals, nutrients, and cancer screening markers.
- Clinical data indicates that over 80% of Function members show at least one biomarker outside the optimal reference range, creating immediate demand for contextualized, patient-specific AI reasoning.
- The release expands on Function’s Private AI Chat developed through its Medical Intelligence Lab™ (MI Lab™), providing users with multi-platform flexibility rather than locking biological data into a single vendor ecosystem.
Grounding Generative AI in Personal Biology
While consumers routinely consult general-purpose AI platforms to decipher symptoms or prepare for clinical consultations, large language models (LLMs) operate without baseline physiological context, defaulting to broad population averages and web search summaries. With over 80% of proactive health consumers presenting at least one biomarker outside optimal clinical ranges (e.g., elevated ApoB, subclinical thyroid dysfunction, or iron deficiency), generic guidance fails to account for individual biological variation.
Function’s interoperability connector replaces manual prompt entry and static PDF uploads with a governed, model-agnostic data pipeline:
- Direct LLM Platform Connectivity: Allows members to authenticate secure data sharing with external frontier models—including Claude (via remote MCP connectors), ChatGPT (via ChatGPT Health and app integrations), and Perplexity—to query longitudinal health trends directly.
- Comprehensive Biomarker Context: Supplies models with structured diagnostic data across more than 160 biannual lab markers spanning cardiovascular risk, metabolic health, hormone regulation, heavy metals, nutrient status, and early oncological risk signals.
- Clinician-Reviewed Annotation Layer: Ingests clinical notes alongside discrete numerical values, equipping AI models with clinical context to distinguish between benign physiological variance and findings requiring formal medical evaluation.
- Ecosystem Independence & Privacy Controls: Built upon Function’s Medical Intelligence Lab™ (MI Lab™) architecture, the connector allows users to grant revocable permissions, maintain ownership over their health records, and switch between foundation models without locking data into a single vendor silo.
“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 now lets you safely share your lab results and clinician-reviewed notes with your preferred AI. You can now choose to truly personalize the conversation by grounding it in your own health.”
— Ziad Sultan, Chief Product and Technology Officer, Function
