AWS Updates: Nov 2026

Amazon Connect Customer now supports Cases and External Tool events in Wait blocks. This allows better control over how customer experience flows respond to real-time data changes. You can wait not only for Lambda function completion but also until Connect case fields meet specific conditions or External Tool calls finish. This enables holding flows until critical information is available, reducing unnecessary transfers and saving both end-user and agent time. For example, you can pause a flow until a refund status field on a case updates, then automatically resume and route the interaction based on the result.

Why it matters

Amazon Connect Customer is an agentic AI solution that helps enterprises deliver exceptional customer experiences. This update is important for improving the efficiency of customer interactions.

Amazon Connect Customer now enables running Model Context Protocol tools within customer experience flows using the new external tools block. This connects to third-party applications and flows via the Amazon Bedrock AgentCore Gateway, allowing agents to retrieve data or trigger actions without custom Lambda functions. For example, a customer inquiring about recent transactions can have the flow pull relevant context from third-party apps like CRM case histories or bank platforms. This feature is available in US East (N. Virginia), US West (Oregon), Canada (Central), Europe (Frankfurt), Europe (London), Asia Pacific (Seoul), Asia Pacific (Singapore), Asia Pacific (Sydney), and Asia Pacific (Tokyo). See the admin guide or Amazon Connect Customer website for details.

Why it matters

This update benefits enterprises aiming to enhance AI-driven customer service experiences by enabling instant retrieval of relevant data during customer interactions, thus streamlining support processes.

Regions: N. Virginia, Oregon and 7 more

AWS Lambda now supports Apache Kafka queues with self-managed Kafka event source mapping (ESM). This allows customers to scale Kafka workloads independent of topic partitions, improving throughput during traffic spikes and processing resilience. Previously, Kafka ESM used the consumer group protocol, assigning each partition to a single consumer, requiring over-partitioning or moving traffic to another queuing system for higher parallelism, which added infrastructure and producer code maintenance overhead. Self-managed Kafka ESM queue support is based on Kafka shared groups, available in general availability with Apache Kafka 4.2. This release enables multiple event pollers to consume from the same partition simultaneously, with each poller supporting up to 10 concurrent invocations, decoupling parallelism from partition count without changing producers, topic settings, or function code. This improves throughput during traffic spikes and allows processing delayed or failed events without delaying other events on the same partition. As queue mode does not preserve partition order, it is suited for workloads where ordering is not required. This feature requires a shared group-enabled Kafka cluster (Apache Kafka 4.2 or later) and is available in all AWS commercial regions where self-managed Kafka ESM is offered. To get started, create a self-managed Kafka ESM with ConsumptionMode set to Queue via the AWS Management Console, Lambda API, AWS CLI, AWS CloudFormation, AWS SAM, or AWS CDK. Standard Lambda pricing applies, and customers configuring provisioning modes for self-managed Kafka ESM also pay for event poller units (EPU). For more information, see the AWS Lambda developer guide and AWS Lambda pricing.

Why it matters

This update is useful for users building event-driven Kafka workloads for use cases such as image and video encoding, fraud detection, notification delivery, and telemetry ingestion.

Amazon Aurora MySQL-Compatible Edition 8.4 now supports MySQL 8.4.10. Key changes include support for GTID tags that allow assigning user-defined tags to GTIDs to identify and group related transactions, and an increase in the maximum concurrent connections per DB instance to 50,000. This enables connection-intensive applications to scale without adding a connection pooling layer. Minor version upgrades can be performed automatically during the maintenance window and can be rolled out gradually across clusters using AWS Organizations' upgrade rollout policy. Upgrades can be executed either in-place or via snapshot restore. This release is supported in all AWS regions where Aurora MySQL is available.

Why it matters

Amazon Aurora is a high-performance, highly available database service compatible with MySQL, designed for global scale

Amazon Aurora MySQL-Compatible Edition 3 (MySQL 8.0-compatible) now supports MySQL 8.0.46, including community MySQL fixes and Aurora-specific enhancements. Aurora MySQL 3.14 adds support for post-quantum TLS (PQ-TLS) key exchange for encrypted data in transit. You can upgrade during a scheduled maintenance window. Enable automatic minor version upgrades and use AWS Organizations' upgrade rollout policy for simplified, phased cluster upgrades, available via in-place or snapshot restore. This release is supported in all AWS regions where Aurora MySQL is available. Aurora offers full MySQL compatibility, global scalability, serverless compute, multi-region resilience, I/O-optimized pricing, built-in security, and continuous backups. For getting started, see the Aurora Getting Started page.

Why it matters

Amazon Aurora is a high-performance, highly available database service compatible with MySQL, and this update introduces new features and security enhancements.

Amazon RDS now supports per-snapshot billing, allowing users to see individual snapshot costs. Previously, snapshot charges were aggregated at the database instance level, hiding details per snapshot. By applying cost allocation tags to snapshots, users can now accurately track backup expenses and manage budgets more effectively. This feature is enabled by default across all RDS engines, commercial AWS regions, and AWS GovCloud (US) regions, at no extra charge.

Why it matters

This update affects users of Amazon RDS, AWS's database service. It enables more precise tracking of backup costs by allowing customers to see the cost of each individual snapshot.

Regions: All Regions

AWS GovCloud (US) now offers the new Anthropic model Haiku 5.5 for Kiro IDE and CLI. The fastest model in the Claude 5.5 family, Haiku 5.5 significantly evolves from Claude Haiku 4.5 and is designed for high-throughput tasks requiring quick responses, enabling developers to summarize, compress, classify and route requests, process text in bulk, and make small edits across many files. It uses adaptive reasoning at five effort levels, defaulting to medium, and is provided with a 1M context window and a 0.06x credit multiplier, while requests exceeding 100K tokens are billed at 0.3x. Update and restart your IDE or CLI to access the new model from the model selector. For details, see GovCloud documentation, monitoring and tracking guides, or contact your AWS account team. Learn more about Kiro on the Kiro product page.

Why it matters

This update provides a new AI model in the Kiro IDE and CLI for AWS GovCloud (US), targeting government and organizations with strict security and compliance requirements.