Millions of Americans Are Already Using AI for Health Questions. Function Lets Members Bring Personal Biology Into the Conversation.

Function launched a secure health data connector that lets members feed 160+ lab results and clinician-reviewed notes into major AI platforms (including ChatGPT, Claude, and Perplexity) to generate biology-informed answers. The company says over 80% of its members have at least one biomarker outside optimal ranges, positioning the rollout as more personalized than population-average AI guidance. Impact is mainly product-level/sector narrative with limited direct financial detail, implying modest immediate market reaction.

Analysis

This is less a monetizable event than a signal that consumer AI is moving one layer deeper into regulated, high-trust workflows. The economic upside accrues first to the platforms that become the default interface for sensitive decisions: more stickiness, more session depth, and a better case for premium subscriptions/enterprise distribution. But that upside is gated by consent friction and liability; if the experience feels even slightly unsafe, usage will stay niche and the revenue line will barely notice.

The more interesting second-order effect is on data infrastructure and compliance tooling. Any workflow that normalizes uploading lab panels and clinician notes to an LLM increases demand for audit trails, access controls, identity verification, and redaction. That is constructive for security vendors and cloud platforms with governance features, but the spend is likely incremental rather than transformative unless this becomes a native feature inside major consumer health apps over the next 6-18 months.

The clearest potential loser is generic health-content/search traffic: if users can get personalized answers without keyword searching, the value of ad-supported symptom content and SEO-driven health portals erodes over time. The counterpoint is that better-informed patients may convert more efficiently into specialist visits and follow-up testing, so the healthcare provider/care-navigation impact is mixed rather than uniformly negative.

Contrarian view: the market may overestimate near-term adoption because the most privacy-sensitive users are also the least willing to connect data to third-party AI. The thesis breaks if regulators or platform policies tighten around health-data sharing, or if engagement data shows that opt-in rates remain low despite the novelty. Until we see persistent usage metrics from the major AI platforms, this reads as a watch item, not a high-conviction thematic trade.

Key Decisions for Investors

• No immediate trade in MSFT/GOOGL on this headline alone; the revenue impact is too small to underwrite a position. Use as a watch item for product-usage data over the next 1-2 quarters.
• Small basket long CRWD/ZS on pullbacks for a 6-12 month horizon if health-data connectors proliferate; thesis is modestly higher demand for consent, identity, and audit controls. Falsify on flat security budgets and no mention of healthcare compliance in vendor commentary.
• Consider a tactical short ZD or ZD put spread over 3-6 months if Similarweb/SEO data starts showing health-content traffic erosion from AI-mediated answers. Risk/reward is asymmetric only if traffic declines are sustained, not one-off.
• Pair trade: long MSFT / short ZD for a 6-12 month thematic hedge. Long leg captures AI-interface stickiness; short leg expresses substitution away from generic health information. Cut if ZD traffic stabilizes or MSFT health engagement fails to inflect.