FranceHealthcare and life sciences2024 – ongoingAlan
Mo AI health assistant with doctor review
French health insurer Alan launched an AI health chatbot whose answers a doctor checks within 15 minutes
Problem
Alan's members could already ask doctors health questions in the insurer's app chat, but having a physician answer every question limits speed and scale. AI chatbots in medicine carry the risk of confident but wrong advice.
Approach
In November 2024 the insurer added Mo to the existing doctor chat. Mo rephrases the member's question, introduces itself and lets the member continue with it or switch to a doctor, so its use is optional. A doctor reviews Mo's answers within 15 minutes and either confirms or corrects them, and the member receives the doctor's note. Before the wider launch Alan ran about 900 test conversations and planned to offer Mo to its 680,000 members.
Results
Takeaway
Alan had doctors review every answer, keeping medical responsibility with physicians while the assistant answered faster. At launch it published no usage, accuracy or cost figures.
Categories
- Industry
- Healthcare and life sciences
- Business function
- Customer service
- Innovation type
- Product and service innovation
- AI technology
- Generative AIConversational and voice AI
- Adoption
- Built in-house
- Evidence
- Company's own claim
Sources
Source summaries
TechCrunch reports the launch of Mo in Alan's chat, describing how members choose between Mo and a doctor and how a doctor checks answers within 15 minutes. It cites about 900 conversations in the preceding weeks and 680,000 covered members, and quotes Alan's AI lead on plans to make Mo a health companion that knows the member's history. It notes the risk of hallucination in medical AI.
HLTH reports that Mo would be offered to 680,000 members after 900 successful test conversations. Physicians check its responses within 15 minutes.
Full details are in the original sources.
