Starting today, Amazon EC2 G7e instances accelerated by NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs are now available in US West (Oregon) region. G7e instances offer up to 2.3x inference performance compared to G6e.

Customers can use G7e instances to deploy large language models (LLMs), agentic AI models, multimodal generative AI models, and physical AI models. G7e instances offer the highest performance for spatial computing workloads as well as workloads that require both graphics and AI processing capabilities. G7e instances feature up to 8 NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs, with 96 GB of memory per GPU, and 5th Generation Intel Xeon processors. They support up to 192 virtual CPUs (vCPUs) and up to 1600 Gbps of networking bandwidth. G7e instances support NVIDIA GPUDirect Peer to Peer (P2P) that boosts performance for multi-GPU workloads. Multi-GPU G7e instances also support NVIDIA GPUDirect Remote Direct Memory Access (RDMA) with EFA in EC2 UltraClusters, reducing latency for small-scale multi-node workloads.

You can use G7e instances for Amazon EC2 in the following AWS Regions: US West (Oregon), US East (N. Virginia) and US East (Ohio). You can purchase G7e instances as On-Demand Instances, Spot Instances, or as part of Savings Plans.

To get started, visit the AWS Management Console, AWS Command Line Interface (CLI), and AWS SDKs. To learn more, visit G7e instances.

​Starting today, Amazon EC2 G7e instances accelerated by NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs are now available in US West (Oregon) region. G7e instances offer up to 2.3x inference performance compared to G6e.
Customers can use G7e instances to deploy large language models (LLMs), agentic AI models, multimodal generative AI models, and physical AI models. G7e instances offer the highest performance for spatial computing workloads as well as workloads that require both graphics and AI processing capabilities. G7e instances feature up to 8 NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs, with 96 GB of memory per GPU, and 5th Generation Intel Xeon processors. They support up to 192 virtual CPUs (vCPUs) and up to 1600 Gbps of networking bandwidth. G7e instances support NVIDIA GPUDirect Peer to Peer (P2P) that boosts performance for multi-GPU workloads. Multi-GPU G7e instances also support NVIDIA GPUDirect Remote Direct Memory Access (RDMA) with EFA in EC2 UltraClusters, reducing latency for small-scale multi-node workloads.
You can use G7e instances for Amazon EC2 in the following AWS Regions: US West (Oregon), US East (N. Virginia) and US East (Ohio). You can purchase G7e instances as On-Demand Instances, Spot Instances, or as part of Savings Plans.
To get started, visit the AWS Management Console, AWS Command Line Interface (CLI), and AWS SDKs. To learn more, visit G7e instances.  

Amazon Bedrock now supports structured outputs, a capability that provides consistent, machine-readable responses from foundation models that adhere to your defined JSON schemas. Instead of prompting for valid JSON and adding extra checks in your application, you can specify the format you want and receive responses that match it—making production workflows more predictable and resilient.

Structured outputs helps with common production tasks such as extracting key fields and powering workflows that use APIs or tools, where small formatting errors can break downstream systems. By ensuring schema compliance, it reduces the need for custom validation logic and lowers operational overhead through fewer failed requests and retries—so you can confidently deploy AI applications that require predictable, machine-readable outputs. You can use structured outputs in two ways: define a JSON schema that describes the response format you want, or use strict tool definitions to ensure a model’s tool calls match your specifications.

Structured outputs is generally available for Anthropic Claude 4.5 models and select open-weight models across the Converse, ConverseStream, InvokeModel, and InvokeModelWithResponseStream APIs in all commercial AWS Regions where Amazon Bedrock is supported. To learn more about structured outputs and the supported models, visit the Amazon Bedrock documentation.

​Amazon Bedrock now supports structured outputs, a capability that provides consistent, machine-readable responses from foundation models that adhere to your defined JSON schemas. Instead of prompting for valid JSON and adding extra checks in your application, you can specify the format you want and receive responses that match it—making production workflows more predictable and resilient. Structured outputs helps with common production tasks such as extracting key fields and powering workflows that use APIs or tools, where small formatting errors can break downstream systems. By ensuring schema compliance, it reduces the need for custom validation logic and lowers operational overhead through fewer failed requests and retries—so you can confidently deploy AI applications that require predictable, machine-readable outputs. You can use structured outputs in two ways: define a JSON schema that describes the response format you want, or use strict tool definitions to ensure a model’s tool calls match your specifications. Structured outputs is generally available for Anthropic Claude 4.5 models and select open-weight models across the Converse, ConverseStream, InvokeModel, and InvokeModelWithResponseStream APIs in all commercial AWS Regions where Amazon Bedrock is supported. To learn more about structured outputs and the supported models, visit the Amazon Bedrock documentation.  

Amazon SageMaker announces general availability of Data Lineage for Apache Spark jobs executed on Amazon EMR and AWS Glue in SageMaker Unified Studio for IDC based domains. Data Lineage provides you with the information you need to identify the root cause of complex issues and understand the impact of changes.

This feature supports lineage capture of schema and transformations of data assets and columns from Spark executions in EMR-EC2, EMR-Serverless, EMR-EKS, and AWS Glue. You can then explore this lineage visually as a graph in SageMaker Unified Studio or query it using APIs. You can also use lineage to compare transformations across Spark job’s history.

Spark lineage is available in all existing SageMaker Unified Studio regions. For detailed information on how to get started with lineage using these new features, refer to the documentation.

​Amazon SageMaker announces general availability of Data Lineage for Apache Spark jobs executed on Amazon EMR and AWS Glue in SageMaker Unified Studio for IDC based domains. Data Lineage provides you with the information you need to identify the root cause of complex issues and understand the impact of changes. This feature supports lineage capture of schema and transformations of data assets and columns from Spark executions in EMR-EC2, EMR-Serverless, EMR-EKS, and AWS Glue. You can then explore this lineage visually as a graph in SageMaker Unified Studio or query it using APIs. You can also use lineage to compare transformations across Spark job’s history. Spark lineage is available in all existing SageMaker Unified Studio regions. For detailed information on how to get started with lineage using these new features, refer to the documentation.  

Amazon Elastic Container Service (Amazon ECS) announces native support for linear and canary deployment strategies for ECS services using Network Load Balancers (NLB). Now, applications that commonly use NLB, such as those requiring TCP/UDP-based connections, low latency, long-lived connections, or static IP addresses, can take advantage of managed, incremental traffic shifting natively from ECS when rolling out updates.

With this launch, ECS customers using NLB can shift traffic in a controlled manner during deployments, such as moving traffic in increments or starting with a small percentage to validate changes before completing a rollout. These deployment strategies provide additional confidence during updates by allowing teams to observe application behavior at each traffic-shift step, and integrate with Amazon CloudWatch alarms to automatically stop or roll back deployments if issues are detected. This is especially valuable for workloads running behind an NLB, such as online gaming backends, financial transaction systems, and real-time messaging services.

To get started, select your NLB target groups, listener, and preferred deployment strategy in the ECS service configuration using the AWS Management Console, AWS CLI, or Infrastructure-as-Code tools. This can be enabled for both new and existing ECS services in all AWS commercial and AWS GovCloud (US) Regions. For more information, see the documentation for Amazon ECS linear deployments and Amazon ECS canary deployments.

​Amazon Elastic Container Service (Amazon ECS) announces native support for linear and canary deployment strategies for ECS services using Network Load Balancers (NLB). Now, applications that commonly use NLB, such as those requiring TCP/UDP-based connections, low latency, long-lived connections, or static IP addresses, can take advantage of managed, incremental traffic shifting natively from ECS when rolling out updates.
With this launch, ECS customers using NLB can shift traffic in a controlled manner during deployments, such as moving traffic in increments or starting with a small percentage to validate changes before completing a rollout. These deployment strategies provide additional confidence during updates by allowing teams to observe application behavior at each traffic-shift step, and integrate with Amazon CloudWatch alarms to automatically stop or roll back deployments if issues are detected. This is especially valuable for workloads running behind an NLB, such as online gaming backends, financial transaction systems, and real-time messaging services.
To get started, select your NLB target groups, listener, and preferred deployment strategy in the ECS service configuration using the AWS Management Console, AWS CLI, or Infrastructure-as-Code tools. This can be enabled for both new and existing ECS services in all AWS commercial and AWS GovCloud (US) Regions. For more information, see the documentation for Amazon ECS linear deployments and Amazon ECS canary deployments.  

• Categoría: Noticias de Microsoft

febrero 4, 2026

Abrir mercados digitales para que la IA pueda comprar —y negociar— por ustedes

Por: Samantha Kubota, escritora de Microsoft.

Imaginen un mundo en el que tienen un asistente digital que puede hacer más que responder a sus preguntas en un chat. En este futuro, podrían enviar a su asistente a un mercado digital para que haga la compra, reserve un vuelo o incluso negocie los términos del contrato de alquiler de su apartamento.

Estos agentes impulsados por IA podrían interactuar con agentes de empresas en su nombre y defenderlos, todo ello sin que tengan que mover un dedo.

Ese futuro no es solo para el mundo de la ciencia ficción. En un artículo publicado el jueves en Communications of the ACM, investigadores de Microsoft afirman que este tipo de economía agéntica abierta es la forma más beneficiosa para que la IA avance, al maximizar las oportunidades tanto para empresas como para particulares.

¿Qué es una economía agente abierta?

En la actualidad, a medida que la economía agéntica toma forma, muchos de los agentes de IA con los que interactuamos existen en lo que se llama un «jardín amurallado». Si necesitan ayuda para reprogramar un vuelo, por ejemplo, podrían hablar con el agente de IA en la web de la aerolínea. Pero en una economía agéntica abierta, dicen los investigadores, habría una «red de agentes» que formaría un ecosistema descentralizado donde los agentes de IA podrían interactuar con libertad entre sí, sin estar confinados al «jardín amurallado» de un solo sitio web.

Comparan la economía agéntica abierta con la promesa de la World Wide Web, en la que cualquier agente podría realizar transacciones con cualquier otro. Los agentes asistentes desempeñarían un papel similar al de los navegadores web, y los agentes de servicio similares a los sitios web.

Uno de los autores del artículo, el investigador de Microsoft David Rothschild, dice que él y sus colegas eligieron escribir sobre este tema porque temen que los «jardines amurallados» aparezcan en algunas plataformas importantes y puedan frenar la innovación. Rothschild señala que algunas empresas ya han construido grandes plataformas en las que querrán mantener a los usuarios existentes.

«Hay muy pocas empresas que hayan capturado todo nuestro tiempo digital», explica. «Tienen un porcentaje enorme de nuestra atención, y harán todo lo posible para mantenerlos aislados en sus plataformas.»

Rothschild dice que iniciar ahora la discusión sobre una economía agente abierta es fundamental.

«Si no, el impulso y la facilidad nos van a empujar a esa versión de jardines amurallados», dice. «Y eso significaría menos bienestar general y menos oportunidades para la sociedad.»

¿Por qué tener una economía agéntica abierta?

En el artículo, investigadores de Microsoft afirman que permitir que muchos agentes de IA diferentes operen en un mercado abierto es el mejor camino para seguir. En este contexto, los agentes ayudan a que los mercados funcionen de forma más fluida, facilitan que las personas cambien entre servicios y dan acceso a más personas a herramientas y servicios digitales de forma descentralizada.

El coautor y director de programa Matt Vogel afirma que cree que una economía agéntica abierta será beneficiosa tanto para las personas como para las empresas.

«Los consumidores pueden encontrar el negocio que mejor se adapte a sus necesidades y cambiar con facilidad entre ellos», dice Vogel. «No están atrapados ni encerrados en uno de ellos.»

Y las empresas pasarán de intentar presentar su producto a través de la publicidad a mejorarlo, dice Rothchild.

«Prevemos un movimiento desde lo que llamaremos la ‘economía de la atención’ hacia la ‘economía de preferencias’», explica. «La esperanza es que las marcas continúen con el gasto de dinero para hacerlas más exitosas, pero lo hacen de una manera que en verdad mejora el producto y mejora nuestra comprensión del producto, en lugar de tan solo ponerse delante de nosotros.»

Cómo llegar a una economía agéntica abierta

Los investigadores afirman que para formar una economía agéntica abierta serán necesarios varios desarrollos tecnológicos y estructurales clave. Primero, la adopción generalizada de agentes asistentes y de servicio, así como la comunicación programática entre agentes. Una vez que las personas y las empresas adopten el uso de agentes de IA como sus representantes, estos deben ser capaces de comunicarse entre sí, de forma no guionizada.

Esto se haría en un mercado digital neutral, similar a cómo se creó la Estación Espacial Internacional en cooperación con empresas y gobiernos, en lugar de en una plataforma única propiedad de una sola empresa.

Para que los agentes puedan hacer su trabajo, los investigadores afirman que sería necesario establecer marcos y protocolos estandarizados para que los agentes puedan descubrirse entre sí y realizar interacciones seguras.

Aunque los consumidores adoptarían y formarían a sus agentes asistentes personales de IA, los investigadores afirman que varios grupos deben participar en la construcción y gestión de la economía agente abierta: empresas tecnológicas, organismos de normalización, empresas/proveedores de servicios, gobiernos y organismos reguladores. Los investigadores sostienen que esto ayudaría a garantizar que los nuevos mercados se mantengan abiertos, competitivos y seguros.

«Para garantizar un futuro seguro y próspero con un mercado abierto, es necesario desarrollar más tecnología que crea un espacio muy cerrado en el que los agentes estén muy limitados, pero que también añada cierta seguridad y cierto control», dice Rothschild. «Este es un intercambio que creemos importante afrontar de manera abierta — una vía abierta de investigación y una vía abierta de desarrollo. Creemos que merece la pena luchar por un futuro más abierto.»

Descubran más en «La economía agéntica.» Para más información sobre trabajos relacionados, consulten la entrada del blog Microsoft Research en Magentic Marketplace, un entorno de simulación de código abierto para estudiar mercados agénticos.

Imagen principal cortesía de da-kuk/Getty Images.

Etiquetas:

• IA
• Innovación
• Tecnología

The post Abrir mercados digitales para que la IA pueda comprar —y negociar— por ustedes appeared first on Source LATAM.

​The post Abrir mercados digitales para que la IA pueda comprar —y negociar— por ustedes appeared first on Source LATAM.  

Amazon DynamoDB global tables now support replication across multiple AWS accounts. DynamoDB global tables is a fully managed, serverless, multi-Region, and multi-active database used by tens of thousands of customers to power business-critical applications. With this new capability, you can replicate tables across AWS accounts and Regions to improve resiliency, isolate workloads at the account level, and apply distinct security and governance controls.

For multi-account global tables, DynamoDB automatically replicates tables across AWS accounts and Regions. This capability allows you to strengthen fault tolerance and helps ensure applications remain highly available even during account-level disruptions, while allowing customers to align data placement with organizational and security requirements. Multi-account global tables are ideal for customers that adopt multi-account strategies or use AWS Organizations to improve security isolation, enforce data perimeter guardrails, implement disaster recovery (DR), or separate workloads by business unit.

Multi-account global tables is available in all AWS Regions and is billed according to existing global tables pricing.

To get started, see the DynamoDB global tables documentation, and visit the AWS developer guide to learn more about the benefits of using a multi-account strategy for your AWS environment.

​Amazon DynamoDB global tables now support replication across multiple AWS accounts. DynamoDB global tables is a fully managed, serverless, multi-Region, and multi-active database used by tens of thousands of customers to power business-critical applications. With this new capability, you can replicate tables across AWS accounts and Regions to improve resiliency, isolate workloads at the account level, and apply distinct security and governance controls. For multi-account global tables, DynamoDB automatically replicates tables across AWS accounts and Regions. This capability allows you to strengthen fault tolerance and helps ensure applications remain highly available even during account-level disruptions, while allowing customers to align data placement with organizational and security requirements. Multi-account global tables are ideal for customers that adopt multi-account strategies or use AWS Organizations to improve security isolation, enforce data perimeter guardrails, implement disaster recovery (DR), or separate workloads by business unit. Multi-account global tables is available in all AWS Regions and is billed according to existing global tables pricing. To get started, see the DynamoDB global tables documentation, and visit the AWS developer guide to learn more about the benefits of using a multi-account strategy for your AWS environment.  

AWS Marketplace now offers a more localized experience for Europe, Middle East, and Africa (EMEA) customers purchasing Professional Service solutions via AWS EMEA Marketplace Operator.

Customers can now procure Professional Services using localized payment methods and receive invoices from AWS EMEA. This removes previous procurement barriers caused by complex payment remittance processes between different AWS entities, which made it difficult for EMEA customers to purchase Professional Services through AWS Marketplace.

Key benefits include support for SEPA (Single Euro Payment Area) payment methods and invoicing consistency from the same AWS entity covering all AWS Marketplace purchases via AWS EMEA Marketplace Operator. This capability is ideal for EMEA customers purchasing consulting, implementation, or managed services through AWS Marketplace. It also benefits organizations that prefer local payment methods such as SEPA direct debit, want to consolidate AWS and Marketplace billing, or are seeking a simpler procurement experience for Professional Services.

This capability is available for EMEA customers who purchase professional services solutions in AWS Marketplace, with AWS EMEA as the Marketplace Operator. To learn more about purchasing Professional Services products in AWS Marketplace and receive invoices issued by AWS EMEA, visit the AWS Marketplace Buyer Guide and AWS EMEA Marketplace FAQs. For more information on how to add a bank account for SEPA, see Managing Your SEPA Direct Debit Payment Method in the AWS Billing and Cost Management user guide. 

​AWS Marketplace now offers a more localized experience for Europe, Middle East, and Africa (EMEA) customers purchasing Professional Service solutions via AWS EMEA Marketplace Operator. Customers can now procure Professional Services using localized payment methods and receive invoices from AWS EMEA. This removes previous procurement barriers caused by complex payment remittance processes between different AWS entities, which made it difficult for EMEA customers to purchase Professional Services through AWS Marketplace. Key benefits include support for SEPA (Single Euro Payment Area) payment methods and invoicing consistency from the same AWS entity covering all AWS Marketplace purchases via AWS EMEA Marketplace Operator. This capability is ideal for EMEA customers purchasing consulting, implementation, or managed services through AWS Marketplace. It also benefits organizations that prefer local payment methods such as SEPA direct debit, want to consolidate AWS and Marketplace billing, or are seeking a simpler procurement experience for Professional Services. This capability is available for EMEA customers who purchase professional services solutions in AWS Marketplace, with AWS EMEA as the Marketplace Operator. To learn more about purchasing Professional Services products in AWS Marketplace and receive invoices issued by AWS EMEA, visit the AWS Marketplace Buyer Guide and AWS EMEA Marketplace FAQs. For more information on how to add a bank account for SEPA, see Managing Your SEPA Direct Debit Payment Method in the AWS Billing and Cost Management user guide.   

Amazon RDS now provides an enhanced console experience that consolidates and provides all relevant information needed to connect to a database in one place, making it easier to connect to your RDS databases.

The new console experience provides ready-made code snippets for Java, Python, Node.js and other programming languages as well as tools like the psql command line utility. These code snippets are automatically adjusted based on your database’s authentication settings. For example, if your cluster uses IAM authentication, the generated code snippets will use token-based authentication to connect to the database. The console experience also includes integrated CloudShell access, offering the ability to connect to your databases directly from within the RDS console.

This feature is available for Amazon Aurora PostgreSQL, Amazon Aurora MySQL, Amazon RDS for PostgreSQL, Amazon RDS for MySQL, Amazon RDS for MariaDB database engines across all commercial AWS Regions.

Get started with the new console experience for database connectivity through the Amazon RDS Console. To learn more, see the Amazon RDS and Aurora user guide

​Amazon RDS now provides an enhanced console experience that consolidates and provides all relevant information needed to connect to a database in one place, making it easier to connect to your RDS databases. The new console experience provides ready-made code snippets for Java, Python, Node.js and other programming languages as well as tools like the psql command line utility. These code snippets are automatically adjusted based on your database’s authentication settings. For example, if your cluster uses IAM authentication, the generated code snippets will use token-based authentication to connect to the database. The console experience also includes integrated CloudShell access, offering the ability to connect to your databases directly from within the RDS console. This feature is available for Amazon Aurora PostgreSQL, Amazon Aurora MySQL, Amazon RDS for PostgreSQL, Amazon RDS for MySQL, Amazon RDS for MariaDB database engines across all commercial AWS Regions. Get started with the new console experience for database connectivity through the Amazon RDS Console. To learn more, see the Amazon RDS and Aurora user guide  

Amazon Aurora DSQL now supports index creation on NUMERIC data types. With this enhancement, you can now use NUMERIC columns in both primary keys and secondary indexes, which can improve query performance for workloads that rely on high-precision values such as currency amounts, measurements and statistical data.

​Amazon Aurora DSQL now supports index creation on NUMERIC data types. With this enhancement, you can now use NUMERIC columns in both primary keys and secondary indexes, which can improve query performance for workloads that rely on high-precision values such as currency amounts, measurements and statistical data.  

Quick Sight in Amazon Quick Suite now enables authors to resolve or update ambiguous locations directly on map visuals for accurate geographical data visualization. When Quick Suite encounters location names that exist in multiple regions—such as cities like Springfield or Abbeville that appear in multiple U.S. states—users can now explicitly define the correct geographical context through three resolution methods: adding supporting geospatial fields to create location hierarchies, searching for specific locations from Quick Suite’s geographical database, or entering exact latitude and longitude coordinates for precise positioning.

These enhancements address visualization accuracy needs for organizations working with datasets containing common location names that could refer to multiple places. With location mapping, dashboard authors can ensure their geospatial visuals correctly represent their data’s geographical context, leading to more reliable insights. The feature provides clear status tracking with Unmatched, Matched, and Unused location indicators, helping users understand and manage their location mappings effectively. Users can resolve ambiguous locations directly from map visuals by selecting «Resolve now» or accessing «Geo data match» options.

This feature is now available in all Amazon Quick Suite regions where Quick Sight is supported. Discover how to create maps and geospatial charts in Quick Suite and learn more about this new feature in our blog post.

​Quick Sight in Amazon Quick Suite now enables authors to resolve or update ambiguous locations directly on map visuals for accurate geographical data visualization. When Quick Suite encounters location names that exist in multiple regions—such as cities like Springfield or Abbeville that appear in multiple U.S. states—users can now explicitly define the correct geographical context through three resolution methods: adding supporting geospatial fields to create location hierarchies, searching for specific locations from Quick Suite’s geographical database, or entering exact latitude and longitude coordinates for precise positioning. These enhancements address visualization accuracy needs for organizations working with datasets containing common location names that could refer to multiple places. With location mapping, dashboard authors can ensure their geospatial visuals correctly represent their data’s geographical context, leading to more reliable insights. The feature provides clear status tracking with Unmatched, Matched, and Unused location indicators, helping users understand and manage their location mappings effectively. Users can resolve ambiguous locations directly from map visuals by selecting «Resolve now» or accessing «Geo data match» options. This feature is now available in all Amazon Quick Suite regions where Quick Sight is supported. Discover how to create maps and geospatial charts in Quick Suite and learn more about this new feature in our blog post.