Announcing Amazon Textract Insights: Verifiable generative AI for documents
Announcing the general availability of Amazon Textract Insights, a new generative AI feature of Amazon Textract. Describe the fields you need in natural language, submit a document or a mixed packet, and receive structured results with a confidence score and page-level grounding for each value, with no model training required.
Amazon Textract Insights, a new generative AI feature of Amazon Textract, has been announced for general availability. This feature allows organizations to describe the fields they need in natural language and receive structured results from complex documents, complete with confidence scores and page-level grounding. Traditional approaches to extracting data from mixed packets, such as custom pipelines or general-purpose language models, often result in trade-offs with limited automation.
Amazon Textract Insights simplifies this process by providing a single managed API, enabling teams to directly work on the extracted data. With Amazon Textract Insights, users define what to extract in a blueprint, which can be a public blueprint for common document types or a custom blueprint for business-specific documents. The service splits mixed packets into logical sub-documents, matches each one to the applicable blueprint, and extracts the requested fields, returning a confidence score and grounding for each value.
This provides auditable automation, as each value is traceable to its source. Vijay Agrawal, Vice President of Information Systems at Symbeo, a CorVel Company, expressed excitement about the potential of Amazon Textract Insights, highlighting its ability to handle diverse document layouts and formats, reducing the need for manual review.
Written by urgent.news from AWS Machine Learning's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.