Details
Watsonx Assistant uses natural language understanding and IBM's large language models, including Granite, to interpret customer queries and deliver context-aware responses. Businesses can deploy branded assistants in existing websites, apps, or phone systems without writing code. A conversational search capability, using retrieval-augmented generation (RAG), allows the assistant to answer policy-grounded questions with little configuration. IBM also introduced purpose-built large speech models (LSMs) for telephone channels to improve voice recognition accuracy.
Have evidence about IBM's AI practices? Submit a report.
Report a Sighting →