MedLens

Company: MedLens

Self-reported purpose: A consumer health-literacy app for U.S. patients and caregivers that uses GPT-5.6 to simplify FDA medicine labels into concise visual facts with traceable citations.

What changed: The project is a self-contained prototype built in a hackathon context, not yet a product in production or with customers.

Single most important open question: Does the author’s approach to validation and evidence handling (e.g., withholding unsupported claims) actually work at scale, or does it fail under real-world edge cases?

The description states that MedLens is a consumer health-literacy app for U.S. patients and caregivers. It allows users to search for a medicine by brand or generic name, select the exact product type, route, dosage form, and FDA label, and then receive a five-part dashboard (Uses; Warnings; Do not use; Side effects; Interactions).

Each visual fact includes:

• A numbered citation.
• A link to the exact, unchanged FDA passage supporting that fact.
• Access to the complete FDA/openFDA record and corresponding NIH DailyMed label.

The app does not provide diagnosis, dosing recommendations, personalized advice, or pill identification.

Inference: MedLens is a tool for retrieval and simplification of official medical information, not an AI-assisted decision-making system. It is positioned as a source-first, evidence-traceable interface to FDA labels.

The description states that the app was built with the goal of making medicine labels easier to navigate without replacing the official source or allowing AI to invent medical claims.

It emphasizes:

• The use of GPT-5.6 for structured extraction and review.
• A structured output approach that constrains response format.
• An independent review process that checks whether each simplification is directly supported by its quoted evidence.
• Unsupported claims are withheld, not filled from model memory.

Inference: The positioning evolved from a general idea of “AI-powered health literacy” to a more specific, source-first, safety-conscious approach. It avoids the common AI trap of overpromising or creating false certainty.

The description states that MedLens is for:

• U.S. patients and caregivers
• Specifically, those seeking clarity on medicine labels

It does not define a细分 customer segment beyond this general audience.

Inference: The ICP appears to be health-literate individuals in the U.S. who are navigating complex or unfamiliar medicine labels, especially when dealing with prescription or OTC drugs.

The description does not state:

• Any pricing model
• Revenue streams
• Monetization strategy
• Customer acquisition plans

Not evidenced: No indication of how the product would be monetized, if at all.

The app was built using:

• Codex
• GPT-5.6
• React, TypeScript, Node.js
• openfda-api, openfda-api, nih-dailymed, responses-api, vinext, vite

Key technical features include:

• Structured extraction of facts from FDA labels.
• Independent review of whether each simplification is supported by evidence.
• Use of Structured Outputs to constrain model responses.
• Deterministic checks via application code to validate claims.
• Visual dashboard with citations and links to original sources.

Inference: The technical approach shows a strong emphasis on validation, traceability, and safety, rather than generative AI overreach. It is built for reliability and compliance, not speed or scale.

The description states:

• The project was submitted to the OpenAI 2026 hackathon
• It is a prototype, not yet in production
• No revenue, customers, or adoption data are provided

Not evidenced: No signs of traction, user base, or product-market fit beyond the author’s own account.

The description does not mention:

• Competitors
• Existing solutions in the health-literacy or medicine-label space
• Market positioning relative to other tools

Not evidenced: No competitive landscape or differentiation analysis is provided.

• Single-person team: The project was built by one person, which raises questions about scalability and long-term maintenance.
• Limited scope of validation: While the app uses structured outputs and checks, it is unclear how well this approach scales to all FDA label edge cases or handles real-world usage.
• No monetization strategy: No indication of how the product would be monetized or sustained.
• Hackathon prototype: The project is a hackathon submission, not a mature product, so its readiness for market is unknown.

1. What are the specific edge cases that have been encountered in real-world FDA label data and how were they handled?
2. How does the app handle discrepancies between different versions of the same label or conflicting information across sources?
3. Has there been any user testing with actual patients or caregivers, and what feedback was received?
4. What are the technical limitations of using GPT-5.6 for structured extraction at scale?
5. Is there a plan to expand beyond U.S. labels or into other languages (e.g., Spanish)?
6. How is the app intended to be monetized or sustained long-term?