用 Amazon Bedrock Data Automation 构建无服务器 PII 脱敏流水线
原标题：Build a serverless PII redaction pipeline with Amazon Bedrock Data Automation
AI 摘要
AWS 机器学习博客发布了一篇技术文章，介绍如何利用 Amazon Bedrock Data Automation（BDA）构建无服务器 PII 脱敏流水线。方案通过自定义 blueprint 以自然语言指令声明需脱敏的字段，结合 Step Functions 与 Lambda 实现批量文档和图像的端到端 PII 检测与脱敏。文章以 Attending Physician Statements 为例，展示了如何区分患者与医生的姓名、日期等字段，并输出置信度与边界框坐标用于后续黑框遮盖。
正文节选
Build a serverless PII redaction pipeline with Amazon Bedrock Data Automation Organizations that process thousands of scanned documents daily, including medical forms, insurance claims, and financial records, face a recurring compliance need: personally identifiable information (PII) redaction before documents are shared with third parties or processed downstream. Manual redaction doesn’t scale: It consumes staff hours, introduces human error, and creates compliance exposure. Redaction is also a