- Alibaba’s AI assistant adds medical and nutrition data sources as it builds out its health offering
- The move expands Qwen’s health capabilities beyond drug information into broader medical knowledge and nutrition data
Alibaba’s Qwen is expanding its health capabilities by adding professional medical and nutrition databases, as the Chinese AI assistant seeks to build a more comprehensive knowledge base for health-related queries.
Qwen said on September 17 that it had added access to the People’s Medical Publishing House’s medical knowledge database and the NutriData nutrition database, extending its health capabilities into medical knowledge and nutritional data.
The medical database, developed by People’s Medical Publishing House, contains structured digital versions of medical textbooks, specialist books and journals accumulated over more than 70 years.
It provides Qwen with a continuously updated knowledge base from a major Chinese medical publisher.
Access to nutrition data
NutriData, jointly developed with the Chinese Nutrition Society and based on authoritative sources including the China Food Composition Tables, covers nutritional components such as calories, protein, fat, vitamins and minerals.
The databases are designed to give Qwen more detailed answers to health-related questions.
For example, when asked about the difference between glycated hemoglobin and fasting blood glucose, Qwen can draw on medical knowledge to explain what each indicator measures and why they are often considered together.
For a question such as whether soy milk or cow’s milk is more suitable for breakfast, it can compare nutritional components such as protein and calcium and take into account information provided by the user, such as lactose tolerance.
Building a professional health base
Qwen has been expanding its health capabilities for several months.
In May, it partnered with the information center of China’s National Medical Products Administration to gain access to millions of records covering drugs, cosmetics and medical devices.
The latest additions broaden that effort from drug and product information to medical knowledge and nutrition data, giving Qwen three major sources of professional health information: regulatory data, medical literature and nutrition databases.
The approach also reflects a broader question facing consumer AI health tools: how to make AI-generated health information more trustworthy without turning the assistant into a substitute for a doctor.
Contrast with the US model
In the US, many AI health products focus on enterprise applications such as clinical support and insurance cost management, while consumer-facing tools have generally been more cautious about providing specific health recommendations.
Qwen is taking a more direct consumer approach, using authoritative databases to underpin its health answers.
Whether that model can be replicated overseas will depend partly on whether other markets have similarly authoritative data sources that are available for AI applications.
For Chinese AI health products, however, Qwen’s strategy points to a model of building a professional knowledge base first and expanding consumer use on top of it.
Header image credit: Qwen
