AWS Launches Swift IoT SDK, Nova 2 Sonic Voice Agent, and AI Document Redaction for Healthcare and Banking

AWS announced several updates on June 24, 2026, including the general availability of the AWS IoT Device SDK for Swift, a healthcare appointment voice agent built with Amazon Nova 2 Sonic, and a scalable document redaction solution for Huntington Bank. The releases target developers building IoT applications on Apple platforms, healthcare automation, and large-scale data privacy.

The AWS IoT Device SDK for Swift is now generally available, enabling Swift developers to build secure, scalable IoT applications natively on macOS, iOS, tvOS, and Linux. This SDK fills a gap in native Swift support for AWS IoT services, providing production-ready APIs for device management and secure communication.

Swift IoT SDK and Healthcare Voice Agent

The SDK includes integrated service clients for AWS IoT Device Shadow, Jobs, and Fleet Provisioning, allowing developers to synchronize device states, manage remote operations, and automate certificate and policy creation. It also supports TLS 1.3 on Apple iOS and tvOS platforms for enhanced security. Developers can install the SDK via Swift Package Manager.

Separately, AWS published a guide on building a healthcare appointment agent using Amazon Nova 2 Sonic and Amazon Bedrock AgentCore. The voice agent handles appointment reminders, authenticates patients by voice, and manages confirmations, cancellations, or rescheduling. It also collects pre-visit health information and escalates to human staff when needed. The sample includes a browser-based interface for testing.

• The agent aims to reduce no-show rates by handling routine calls at scale.
• It uses natural language processing for patient interaction.
• Integration with phone lines is left for further development.

Document Redaction and BI Integration

Huntington Bank built a scalable AWS solution to detect and redact Personally Identifiable Information (PII) and Payment Card Industry (PCI) data from over 400 million documents. The system reduced processing time from years to a few months, achieving over 95% redaction accuracy. The solution uses AWS machine learning services for automated data classification and redaction.

Additionally, AWS published a guide on integrating Snowflake semantic views with Amazon Quick for AI-powered business intelligence. The integration allows users to load data from Amazon S3 into Snowflake, define semantic views, and query them using natural language through Cortex Analyst. The resulting datasets can be used to create Amazon Quick dashboards, enabling BI teams to ask natural-language questions.

These updates reflect AWS's focus on expanding developer tools for Apple platforms, automating healthcare workflows, and improving data privacy and analytics for enterprises.