Singapore Airlines doubles chatbot CSAT using OpenAI API
·2026.11.10 17:00
Key point
The airline cut agent handling time by 10% and tripled Flight Recommender usage by integrating OpenAI models into customer support and internal workflows.
Details
- Singapore Airlines integrated OpenAI API and models into customer support, travel discovery, and internal operations, resulting in a 2x increase in customer satisfaction score (CSAT) for the Kris chatbot.
- Customer service representatives using AI-assisted tools, including call summarization, saw an approximately 10% reduction in average handling time.
- Flight Recommender, a natural language travel discovery tool built on the OpenAI API, experienced nearly 3x growth in monthly users over the past year.
- Internal software development teams achieved productivity gains of up to 20% by using OpenAI models for requirements refinement, test case creation, and code generation.
- Over 60% of employees now use Jarvis, the airline's internal AI platform powered by OpenAI models, for document summarization and custom workflows.
- The Kris chatbot transitioned from a scripted dialog-flow system to an LLM-powered conversational experience, allowing it to interpret natural language intent and manage multi-topic conversations.
- Future expansions include deeper transactional capabilities for Kris, such as booking changes and baggage transactions, and a more conversational interface for Flight Recommender.
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.