Amazon launches Agentic AI platform on Amazon Bedrock

Amazon launches Agentic AI platform on Amazon Bedrock. The solution offers governance and security tools to build agents efficiently. Available only in the eu-west-1 region.

Amazon is launching an Agentic AI platform on Amazon Bedrock to make it easier to build and manage agents securely and efficiently. The solution offers governance and security tools, allowing teams to operate autonomously within defined limits, with centralized visibility and governance. The platform uses a federated model, in which the platform team offers governed services and the workstream teams develop and operate agents.

This reduces the complexity of managing agents while maintaining control over data and processes. Components such as inference gateways, access control, security policies and cost management are integrated, ensuring that the solution is scalable and secure. Amazon also highlighted the ability to federate data from different sources without moving data between systems, which is essential for companies that need to work with scattered information.

The platform includes the architecture of the Adaptive Agentic Intelligence Framework (AAIF), which uses small models (SLMs) for local tasks and cloud models via Amazon Bedrock for complex tasks. Bayer Crop Science developed the Agentic Decision Engine (ADE), a natural-language-based interface on top of the Connected Decision Engine (CDE), which cuts the time needed to diagnose workflow failures from 1 hour to less than 10 minutes. The solution is based on technologies such as Amazon Bedrock AgentCore and includes implementation examples from companies such as Bayer Crop Science, Octave and Denodo. The platform is available only in the eu-west-1 region, with the date and availability in Brazil not disclosed.

Reduced diagnostic time and data federation

Reduced diagnosis time and data federation allow teams to identify and resolve problems without interrupting operations Reduced time is crucial for sectors such as finance and healthcare, where latency can affect critical decisions. This feature is especially useful for companies that have data scattered across multiple platforms, avoiding the need to transfer large volumes of data. The platform's layered architecture also offers greater flexibility, allowing teams to adapt the solution to their specific needs. This structure helps maintain scalability, which is important for companies looking to expand their automated processes. Amazon's Agentic AI platform changes the way agents are developed, using federated models and governance tools to run systems securely and at scale. This reduces the need to move data between systems, maintaining privacy and security throughout the entire process.

Text written with AI from the sources listed and checked by the newsroom's automated rules. How we work

• #hardware
• #software
• #ia