SNCB Launches AI Assistant: Boosts Passenger Query Accuracy to 99%+
SNCB’s Mobi 2.0 AI assistant boosts passenger support, understanding questions with over 99% accuracy. This improves efficiency, freeing staff for complex issues.

- Belgian rail operator SNCB has deployed Mobi 2.0, a generative AI virtual assistant to handle passenger inquiries.
- The system improves question understanding from 65% in the previous version to over 99%, across a volume of ~30,000 monthly conversations.
- This move positions SNCB as an early public-sector adopter of GenAI for customer service, aiming to improve efficiency and reallocate human agents to more complex issues.
BRUSSELS, BELGIUM – Belgian national rail operator SNCB has launched Mobi 2.0, a generative AI-powered virtual assistant, across its website and messaging channels. The new system is designed to handle a high volume of passenger queries with significantly improved accuracy, freeing up customer service staff for more complex requests. The cost of the project was not disclosed.
| Category | Specification / Detail |
|---|---|
| Project / Model | Mobi 2.0 Virtual Passenger Assistant |
| Operator | SNCB (Belgian National Railway Company) |
| Technology | Generative Artificial Intelligence |
| Question Understanding Rate | >99% (up from ~65% in previous version) |
| Current Monthly Volume | ~30,000 conversations |
| Platform Availability | SNCB Website, WhatsApp, Messenger (Mobile App integration pending 2024) |
| Languages Supported | French, Dutch, German, English |
Operational & Technical Details
Mobi 2.0 represents a significant performance upgrade over SNCB’s previous assistant, in use since 2020. The new model’s question comprehension rate exceeds 99%, a substantial increase from the prior 65% benchmark. This allows it to process complex sentences and user spelling errors more effectively.
The system operates 24/7, fielding frequently asked questions about routes, fares, and delays. To ensure data security and response accuracy, Mobi 2.0 functions in a closed digital environment. It is fed exclusively with SNCB’s internal commercial information. For issues requiring human intervention, such as refund requests, the assistant facilitates a direct transfer to a live agent.
Market Impact Analysis
SNCB’s deployment makes it one of the first major public service entities in Belgium to adopt generative AI on a large scale for customer interaction. This strategy aligns with a broader industry trend where organizations use AI to manage high-volume, routine inquiries, thereby optimizing human resource allocation. By automating front-line support, SNCB can direct its specialized customer service teams to handle non-standard situations that demand direct human oversight.
The implementation serves as an internal pilot for SNCB, which plans to expand the use of AI technology to other functions and communication channels. This positions the operator as a key test case for AI adoption within Europe’s public transport sector, potentially setting a new standard for passenger service efficiency.
FAQ: Quick Facts
What is the main value of this project?
The project’s primary value is boosting customer service efficiency. It increases the automated question resolution rate from approximately 65% to over 99%, allowing human agents to focus on more complex passenger needs.
When is the system fully operational?
The assistant is currently live on the SNCB website, Messenger, and WhatsApp. Full integration into the SNCB mobile app is scheduled for later this year.



