AWS News - 2026-05-22

Amazon SageMaker Inference now supports OpenAI-compatible APIs, so you can use the tools and frameworks you already know, like the OpenAI SDK, LangChain, and Strands Agents, to connect directly to your SageMaker endpoints. Switching requires nothing more than changing an endpoint URL — no custom integration code, no SDK wrappers, no rewrites.

With this launch, you no longer need to adopt a different API format or change your authentication approach. Simply change your endpoint URL, and your existing SDK calls, streaming logic, and framework integrations continue to work as-is. You immediately gain the ability to choose your own GPU instances, keep data in your own VPC, run any open source or fine-tuned model, and scale with auto-scaling policies tuned to your workload. Authentication uses existing AWS credentials with automatic token refresh, so there is nothing extra to manage in production.

This capability is available today in US East (N. Virginia), US West (Oregon), US East (Ohio), Asia Pacific (Mumbai), Asia Pacific (Jakarta), Europe (Ireland), Europe (Frankfurt), South America (São Paulo), Asia Pacific (Tokyo), Asia Pacific (Seoul), Europe (London), Asia Pacific (Singapore), Asia Pacific (Sydney), and Canada (Central). To learn more and get started, read the launch blog or visit the SageMaker Inference documentation.

Amazon SageMaker Unified Studio now supports automatic creation of connections for Glue job retries across subnets to improve data pipeline resilience. This helps organizations running business-critical data pipelines reduce unplanned downtime and meet their SLAs — without requiring engineers to manually configure backup connectors or intervene during subnet failures.

With this launch, SageMaker Unified Studio automates the provisioning of Glue connectors across subnets defined in the domain VPC configuration. Administrators can define their domain VPC with multiple private subnets across availability zones, and the system provisions the connectors needed for all new projects so that failed jobs can be retried on an alternate subnet automatically. If a Glue job fails because the primary subnet is unavailable due to IP address exhaustion or availability zone degradation, the job can be retried on a connector in a different subnet. No user action is needed beyond the initial VPC configuration on the domain.

This feature is available in all AWS Regions where Amazon SageMaker Unified Studio is available. To learn more, visit the Amazon SageMaker Unified Studio documentation. 

Starting today, Amazon Elastic Compute Cloud (Amazon EC2) C7i-flex, M7i-flex and M7i instances powered by custom 4th Gen Intel Xeon Scalable processors (code-named Sapphire Rapids) are available in Asia Pacific (Hyderabad) region. These custom processors, available only on AWS, offer up to 15% better performance over comparable x86-based Intel processors utilized by other cloud providers.

C7i-flex and M7i-flex instances are the easiest way for you to get price-performance benefits for a majority of general-purpose workloads. They deliver up to 19% better price-performance compared to C6i and M6i  instances respectively. These instances offer the most common sizes, from large to 16xlarge, and are a great first choice for applications that don't fully utilize all compute resources such as web and application servers, virtual-desktops, batch-processing, and microservices. 

 

M7i deliver up to 15% better price-performance compared to M6i. M7i instances are a great choice for workloads that need the largest instance sizes or continuous high CPU usage, such as gaming servers, CPU-based machine learning (ML), and video-streaming. M7i offer larger instance sizes, up to 48xlarge, and two bare metal sizes (metal-24xl, metal-48xl). These bare-metal sizes support built-in Intel accelerators: Data Streaming Accelerator, In-Memory Analytics Accelerator, and QuickAssist Technology that are used to facilitate efficient offload and acceleration of data operations and optimize performance for workloads.

To learn more, visit the EC2  C7i-flex and M7i/M7i-flex instances pages.

Amazon CloudWatch Logs Insights query language now supports 13 new commands and functions that give you more powerful ways to query, transform, and analyze your logs. Customers analyzing logs in CloudWatch Logs Insights often need to perform string manipulation, encode or decode field values, parse non-JSON log formats, or calculate geographic distances, so they can derive deeper insights from their logs.

With this launch, CloudWatch Logs Insights provides new string and numeric functions (round, startswith, endswith, case, regex_replace, haversine), encoding and decoding functions (urlencode, urldecode, base64encode, base64decode), and new parse and analysis commands (parse logfmt, expand, relevantfields). You can now filter logs by string prefixes, decode Base64 payloads inline, parse logfmt structured logs into fields, expand nested JSON arrays into individual records, calculate distances between coordinates, and automatically surface relevant fields in high-cardinality log groups.

These commands and functions are available today in all commercial AWS Regions. To learn more, see the Amazon CloudWatch Logs documentation.

AWS Clean Rooms now supports mutable fine-grained payment configurations for collaboration members. This capability offers customers greater flexibility and control over payment responsibilities as they develop new use cases with their partners. With this launch, customers can specify which partners are authorized to pay for specific cost types after a collaboration is created—including SQL queries, PySpark jobs, ML model training and inference jobs, and synthetic data generation in AWS Clean Rooms.

With AWS Clean Rooms, you can add or remove authorized payers for specific cost types through a change request. Collaboration members must approve the results before it takes effect. Payment configurations support multiple authorized payers for SQL and PySpark analyses. You can select an authorized payer when submitting the analysis. For example, a pharmaceutical research company collaborates with healthcare organizations for real-world clinical trial data. The pharmaceutical research company can pay for complex analysis, and the healthcare organizations can pay for simple SQL analyses in a collaboration. 

AWS Clean Rooms helps companies and their partners easily analyze and collaborate on their collective datasets without revealing or copying one another’s underlying data. For more information about the AWS Regions where AWS Clean Rooms is available, see the AWS Regions table. To learn more about collaborating with AWS Clean Rooms, visit AWS Clean Rooms.

AWS Secrets Manager now extends its managed external secrets capability to include Datadog Keys and Snowflake Programmatic Access Tokens (PATs). Managed external secrets enable customers to automatically rotate third-party credentials directly from AWS Secrets Manager by offering first-class integration with supported third-party services.

With this launch, you can manage rotation for three types of Datadog credentials — API keys, Application keys, and admin credential pairs for service accounts. For Snowflake, you can now rotate Programmatic Access Tokens using Snowflake's native authentication, with a configurable grace period that allows applications to seamlessly transition to new tokens without interruption.

These new integrations join existing managed external secrets integrations with BigID, Confluent Cloud, MongoDB Atlas, and Salesforce, enabling customers to manage third-party software vended secrets.

Datadog and Snowflake PAT managed external secrets are available in all AWS Regions where AWS Secrets Manager managed external secrets is supported. To learn more, visit the AWS Secrets Manager managed external secrets documentation.

AI エージェントに本番データを分析させるには、単にモデルと API をつなぐだけでは足りません。認可、ビジネ […]

本記事は、シリーズ「AWS における AI エージェント対応のデータ基盤」の第 2 回です。第 1 回では、A […]

本記事では、キヤノン IT ソリューションズ株式会社様が 、Amazon Q Developer を開発現場に導入し、3 か月間効果検証を実施した取り組みをご紹介します。コード生成やレビュー支援による効率化、現場での活用事例、そして検証から得られた知見について詳しく解説します。

こんにちは。ソリューションアーキテクトの東 健一です。普段はパブリックセクター技術統括本部で中央省庁のお客様の […]

AWS は、Amazon Automated Reasoning Group、AWS Cryptography、オープンソースコミュニティと協力し、ポスト量子暗号 (PQC) ML-KEM の形式的に検証された最適化実装 mlkem-native を開発しました。本記事では、CBMC によるメモリ安全性・型安全性の検証、HOL Light と s2n-bignum によるアセンブリ実装の正当性証明、SLOTHY によるマイクロアーキテクチャ最適化を組み合わせ、セキュリティ・性能・保守性を同時に実現した取り組みをご紹介します。AWS-LC への統合により、c7i や c7g で約 2 倍の性能向上を達成しました。

AWS Customer Incident Response Team (CIRT) が観測している、攻撃者がお客様アカウントを侵害した後に AWS Organizations から離脱させ、SCP やガバナンス制御を回避する新しい手口について解説します。 具体的には、organizations:LeaveOrganization 権限を持つクレデンシャルが悪用されると、メンバーアカウントが Organization の保護下から外れ、CloudTrail の組織トレイル、GuardDuty の中央集約、SCP による制限、一括請求などの可視性と統制が失われます。 最も効果が大きく労力の少ない対策として、organizations:LeaveOrganization アクションを拒否する SCP (DenyLeaveOrganization) の実装を推奨します。あわせて、CloudTrail での AcceptHandshake / LeaveOrganization / InviteAccountToOrganization / RemoveAccountFromOrganization イベントの監視、IAM の最小権限原則の徹底、およびルートアクセスの一元管理についても解説しています。

Managing identities and access across complex environments has become more critical than ever. AWS Directory Service for Managed Microsoft Active Directory, also known as AWS Managed Microsoft AD, has added new capabilities to manage users and groups. Now, you can perform create, read, update, and delete (CRUD) operations on users and groups directly through AWS […]

We’re excited to announce that Amazon Web Services (AWS) has completed the S&P Global Know Your Third Party (KY3P) assessment of its security posture. This assessment demonstrates our continued commitment to meet the heightened expectations of cloud service providers. Customers can now use the AWS KY3P assessment to reduce their supplier due diligence burden. KY3P, […]

This solution combines the power of Amazon Bedrock AgentCore, Strands Agents, and Amazon Quick transforms to deliver a secure, scalable, and intelligent system for building and operating AI agents while transforming data into actionable business insights.

In this post, we demonstrate how to build an AI-powered recruitment assistant using Amazon Bedrock that brings efficiencies to candidate evaluation, generates personalized interview questions, and provides data-driven insights for human hiring decisions. This post presents a reference architecture for learning purposes — not a production-ready solution. Amazon Bedrock and the AWS services used here are general-purpose tools that customers can combine to support a wide variety of use cases, including recruitment workflows. The architecture demonstrates one possible approach; customers should adapt it to their specific requirements.

In this post, we show you how OPLOG developed three AI agents using the Strands Agents SDK, deployed them to Amazon Bedrock AgentCore, and integrated Amazon Bedrock with Anthropic’s Claude Sonnet and Amazon Bedrock Knowledge Bases for Retrieval Augmented Generation (RAG).

In this post, you will learn how to implement Recursive Language Models (RLM) using Amazon Bedrock AgentCore Code Interpreter and the Strands Agents SDK. By the end, you will know how to process documents of varying lengths, with no upper bound on context size, use Bedrock AgentCore Code Interpreter as persistent working memory for iterative document analysis, and orchestrate sub-large language model (sub-LLM) calls from within a sandboxed Python environment to analyze specific document sections.

This post explores design considerations for architecting multi-tenant agentic applications and the framework needed to address SaaS architecture challenges with Amazon Bedrock AgentCore.

This post shows you how to use Amazon Bedrock AgentCore Runtime with Model Context Protocol (MCP) support to connect Amazon Quick with AWS services through the AWS API MCP Server, creating a conversational AI assistant that translates natural language into AWS Command Line Interface (AWS CLI) commands, without the need to switch between tools during critical moments.

Many healthcare organizations report that traditional worklist systems rely on rigid rules that ignore critical context, radiologist specialization, current workload, fatigue levels, and case complexity. This creates a persistent challenge: radiologists cherry-pick easier, higher-value cases while avoiding complex studies, leading to diagnostic delays and increased costs. Research across 62 hospitals analyzing 2.2 million studies found […]

In this post, you will learn what Nova Act offers, how HIPAA eligibility applies to agentic AI, and how to get started.