The feature processes mixed packets containing up to 3,000 pages submitted from an Amazon Simple Storage Service bucket in a single asynchronous request. Target fields are defined through JSON Schema blueprints, which can be selected from standard public templates or custom-authored by organizations. The tool splits diverse packets into logical sub-documents to capture both verbatim text and contextually inferred values. Extracted results are written back to storage with exact bounding-box coordinates to facilitate enterprise verification.
Key points
Enterprises can extract structured data from complex files using natural language instruc…: Enterprises can extract structured data from complex files using natural language instructions without conducting model training.
The service automatically divides mixed document bundles into logical sub-documents and m…: The service automatically divides mixed document bundles into logical sub-documents and maps them to appropriate schema blueprints.
Workflows support verifiable automation by returning auditable location coordinates along…: Workflows support verifiable automation by returning auditable location coordinates alongside both verbatim and inferred data points.
Sources checked by our analysis pipeline: 1
Artificial Intelligence
Announcing Amazon Textract Insights: Verifiable generative AI for documents
Organizations across financial services, insurance, healthcare, and the public sector depend on document processing. They turn scanned files and paper forms into structured data their systems can act on. As generative AI has matured, these organizations want more than accurate text extraction. They want to describe the information they need in their own words and get dependable, auditable answers from complex documents.
Today, we are announcing the general availability of Amazon Textract Insights, a new generative AI feature of Amazon Textract available through a new API within Amazon Textract. With Amazon Textract Insights, you describe the fields you want in natural language and submit a document or a mixed, multi-document packet. You receive structured results in which each value carries a confidence score and page-level grounding, with no model training required.
The challenge of extracting data from mixed packets
Many of the high-value document workloads involve diverse, mixed packets rather than clean, single-purpose files. A mortgage file, for example, typically arrives as a single scan holding a W-2, payslips,…
The prospective deal reflects a rapid expansion of the company's capital base during an intensely active fundraising period this year. If finalized, the transaction would represent a substantial jump in enterprise worth compared to its prior financing marks set just months earlier. This continued influx of capital highlights an aggressive push to secure large-scale resources in 2026.
A controversial terms-of-service revision has sparked significant backlash among veteran merchants who object to their work being utilized for artificial intelligence development. Fearing the loss of control over their creative rights, numerous shop owners are actively closing their accounts in protest. This platform exodus highlights mounting friction between online marketplaces and the creators who supply their digital catalogs.
According to a PBS NewsHour report, President Donald Trump is pushing domestic communities to embrace local data center construction while threatening severe penalties for companies that build overseas. The directive aims to bolster American dominance in superintelligence and prevent China from seizing critical infrastructure. This initiative directly counters recent midterm election results, in which voters across the country passed numerous measures to halt or restrict local facility developments.