American Express Global Business Travel (Amex GBT) has announced new AI-powered customer solutions. Through the use of large language models (LLM), travelers get on-demand support through more intuitive interactions with a virtual agent that understands and responds to complex requests more naturally. In addition, the company has also introduced a natural language query (NLQ) to its data and reporting feature Egencia Analytics Studio, making it easier for travel managers to access travel program data by asking questions in plain language, and go beyond standard reporting to access detailed insights about travel spending and patterns.
The LLM-powered virtual agent, available to Egencia customers, was trained using help center content to generate accurate and contextually appropriate responses. It can handle inquiries in ten different languages including English, French, Spanish and German. This multilingual capability ensures that the virtual agent can assist a diverse range of users from various regions. Questions it can answer include: “How do I reset my password?” or “How do I add my frequent flyer number to my account?”
During testing, the Amex GBT Analytics and Data Science team found the virtual agent increased efficiency and enhanced the customer self-service experience for travelers. Amex GBT data showed that nearly a third of customers who used the virtual agent during a trial phase were able to self-service without proceeding to wait for a travel consultant. The virtual agent experience at Amex GBT was first introduced in 2020 and the addition of LLM supported chat is the most recent evolution of its capabilities.
The NLQ data and reporting feature is designed to make data analysis simpler for travel managers using Egencia. It empowers them with enhanced data analysis capabilities, enabling them to extract key insights and showcase into easy-to-understand visual formats such as charts and graphs. Through a search field in the Analytics Studio dashboard, travel managers can easily locate data insights by asking direct questions such as “What is the quarterly spend for the past three quarters?” The feature allows more than one data metric per question, such as “What is spend, transaction count, average ticket price (ATP) for air in city Paris 2024?” It is currently in beta with plans for a full release slated for 2026.
For more information, visit www.amexglobalbusinesstravel.com.
Related Stories
ASTA Launches New Corporate Initiative
Mesh Payments, Altour Ink New Partnership