Meta Introduces Autodata: An Agentic Framework That Turns AI Models into Autonomous Data Scientists for High-Quality Training Data Creation

Meta AI introduced Autodata, a closed-loop system that uses an AI model as a data scientist to generate, test, and refine datasets.

The framework grounds outputs in source documents and uses specialized subagents to evaluate examples and improve data quality over multiple iterations.

Early results show Autodata outperforming older synthetic data methods on scientific reasoning benchmarks.

The launch highlights a growing idea in AI: spending more inference-time compute can create better training data with less human annotation.