Live intelligence

Latest AI updates

1 updates
Products Editor selected

Amazon Textract Insights launches with natural-language document extraction and page-level grounding

AWS announced general availability of Amazon Textract Insights, a generative AI feature that extracts structured data from documents using natural-language blueprints without model training. It can process single documents or mixed multi-document packets, splitting them, matching sub-documents to blueprints, and returning values with confidence scores and page-level bounding-box grounding. The service supports PDF, PNG, JPEG, and TIFF inputs and documents in English, French, German, Italian, Portuguese, and Spanish, priced per page. AWS also said its Textract Core engine gained accuracy improvements, denser table and handwriting handling, and expanded language coverage.

Why it matters: It provides a concrete managed API for auditable generative AI document extraction, reducing the need to assemble custom OCR, splitting, classification, prompting, and review pipelines.

Amazon Textract Insightsintelligent document processingAWS