ChinaEnergy and utilities2023 – ongoingENN Energy
Large-language-model customer-service chatbot
ENN Energy reported chatbot understanding accuracy of 81.8%, up from 35.7%
Problem
ENN Energy's previous chatbot had limited accuracy in understanding customer requests. Customers also needed access to service outside normal working hours.
Approach
ENN Energy launched a replacement customer-service chatbot based on large language models. It identified customer needs and provided service suggestions around the clock.
Results
Takeaway
The reported improvement measured semantic understanding; ENN Energy did not disclose a corresponding resolution rate or quantified waiting-time reduction.
Categories
- Size
- Large enterprise
- Industry
- Energy and utilities
- Business function
- Customer service
- Innovation type
- Process innovation
- AI technology
- Generative AIConversational and voice AI
- Adoption
- Built in-house
- Evidence
- Company's own claim
Sources
Source summaries
ENN Energy reported launching an LLM-based customer-service chatbot during 2023. It stated that semantic-understanding accuracy increased from 35.7% to 81.8% and that the service operated around the clock.
Full details are in the original sources.
