Pinecone, one of the companies that helped pioneer the vector database market supporting retrieval-augmented generation (RAG), is sending a different message around AI agents. The News Stack reported that Pinecone has provided 800,000 developers with chunking, embedding and retrieval methods for RAG over the past four years and unveiled its agent knowledge engine, Nexus, in early May. Pinecone argues RAG is unstable, slow and costly, but faces questions over industry adoption of its KnowQL.

Dinotesia said on Tuesday it has released a beta version of LegalQ, an AI search service that lets users explore laws and provisions using natural language. The service can find relevant statutes and related clauses without requiring the name of a law or article number. It applies an “agentic RAG” approach that rewrites user intent into legal terms and selects a search method. It uses official data from the Ministry of Government Legislation’s shared law information service.