AWS released Amazon Bedrock Data Automation, a service for automating personally identifiable information (PII) redaction in documents, on a date unspecified, stating it can process thousands of scanned files daily. It is the company's first major update to its machine learning offerings since the launch of Amazon Bedrock in 2023.
Amazon Bedrock Data Automation is built on Amazon Bedrock, targeting compliance needs in industries like healthcare and finance. Availability begins with a public preview, initially for developers and enterprise customers.
"Generative AI document understanding helps remove traditional redaction constraints," said a representative from AWS. "Foundation models can interpret a document page holistically, including its layout, field labels, and context, and can distinguish whose information a field belongs to using plain-language instructions rather than trained entity models."
The announcement follows AWS's broader push into generative AI security and compliance. AWS said the tool helps organizations meet regulatory requirements by automating PII detection and redaction at scale.
AWS did not say how the tool will handle unstructured text in non-English languages, and it raised the open question of how to balance redaction precision with data utility. The company said it plans to expand the tool's capabilities through further experimentation and automation.
Source: awsml
