Tag
This paper introduces AIDA, an autonomous agent framework designed to transform fragmented enterprise data into actionable business insights by leveraging reinforcement learning and a proprietary Domain-Specific Language for SQL execution.
OpenAI details an internal AI data agent built on GPT-5, Codex, and their Evals/Embeddings APIs that lets employees query over 600 petabytes across 70k datasets using natural language. The tool reduces time-to-insight from days to minutes and is used across Engineering, Finance, Research, and Go-To-Market teams.