Reply announced the launch of MLFRAME Reply, a new generative artificial intelligence framework for heterogeneous knowledge bases. Developed by Machine Learning Reply, a specialised company within the group, MLFRAME Reply applies a proprietary methodology to analyze databases, train algorithms, and validate results on top AI technologies. This enables the rapid creation of conversational generative models applicable to specific corporate knowledge domains.

The current availability of generative artificial intelligence technologies makes large amounts of unstructured information and textual data accessible and queryable through natural language. It provides the ability to create distributed and usable knowledge bases both within companies and to be used in customer relations. MLFRAME Reply acts as an engine that extracts knowledge through natural language, reaggregates it, and redistributes it in conversational form.

It empowers the 'artificial intelligence' component underlying the new generation of 'human-like' interaction systems, such as digital assistants and digital humans. MLFRAME Reply enables businesses to benefit from comprehensive support in all phases of conversational system development and training: from creating a substantial knowledge base within a knowledge domain, to introducing generative models (based on information retrieval and LLM techniques), and training and optimizing algorithms using the most appropriate techniques for the complexity level required for each use case. For example, MLFRAME Reply has already been employed in the development of onboarding systems based on conversational interfaces and has provided conversational intelligence for digital human models in the banking, insurance, and automotive sectors.