Bedrock is a service that enables companies to build generative AI applications through a number of foundation models from leading AI companies and already has tens of thousands of customers.

The upgrades – available in preview - announced in his keynote speech by AWS CEO Matt Garman were:

• Automated Reasoning checks is the first and only generative AI safeguard that helps prevent factual errors due to model hallucinations, opening up new generative AI use cases that demand the highest levels of precision.
• Customers can use multi-agent collaboration to easily build and orchestrate multiple AI agents to solve problems together, expanding the ways customers can apply generative AI to address their most complex use cases.
• Model Distillation empowers customers to transfer specific knowledge from a large, highly capable model to a smaller, more efficient one that can be up to 500% faster and 75% less expensive to run

The new capabilities for Amazon Bedrock will help customers prevent factual errors due to hallucinations, orchestrate multiple AI-powered agents for complex tasks, and create smaller, task-specific models that can perform similarly to a large model at a fraction of the cost and latency.

“With a broad selection of models, leading capabilities that make it easier for developers to incorporate generative AI into their applications, and a commitment to security and privacy, Amazon Bedrock has become essential for customers who want to make generative AI a core part of their applications and businesses,” said Dr Swami Sivasubramanian, vice president of AI and Data at AWS. “That is why we have seen Amazon Bedrock grow its customer base by 4.7x in the past year. Over time, as generative AI transforms more companies and customer experiences, inference will become a core part of every application. With the launch of these new capabilities, we are innovating on behalf of customers to solve some of the top challenges, like hallucinations and cost, that the entire industry is facing when moving generative AI applications to production.”

Prevent factual errors due to hallucinations

Even the most advanced, capable models can provide incorrect or misleading responses. These “hallucinations” remain a fundamental challenge across the industry, limiting the trust companies can place in generative AI, especially in highly regulated industries where accuracy is critical. Automated Reasoning checks is the first and only generative AI safeguard that helps prevent factual errors due to hallucinations using logically accurate and verifiable reasoning. Automated Reasoning checks now enables Amazon Bedrock to validate factual responses for accuracy, produce auditable outputs, and show customers exactly why a model arrived at an outcome.

Easily build and coordinate multiple agents to execute complex workflows

AI-powered agents can help customers’ applications take actions (for example, helping with an order return or analysing customer retention data) by using a model’s reasoning capabilities to break down a task into a series of steps it can execute. While a single agent can be useful, more complex tasks, such as performing financial analysis across hundreds or thousands of different variables, may require a number of agents with their own specialisations. However, creating a system that can co-ordinate multiple agents, share context across them, and dynamically route different tasks to the right agent requires specialised tools and generative AI expertise that many companies do not have. That is why AWS has expanded Amazon Bedrock Agents to support multi-agent collaboration, empowering customers to easily build and co-ordinate specialised agents to execute complex workflows. 

Create smaller, faster, more cost-effective models

Even with the wide variety of models available, it’s still challenging for customers to find one with the right mix of knowledge, cost, and latency that’s best suited to their business needs. Model distillation is a technique that transfers the knowledge from a large model to a small model, while retaining the small model’s performance characteristics. To do this usually requires specialised machine learning (ML) expertise, however with Amazon Bedrock Model Distillation, any customer can now distil their own model with no ML expertise. Customers simply select the best model for a given use case and a smaller model from the same model family that delivers the latency their application requires at the right cost. After the customer provides sample prompts, Amazon Bedrock will do all the work to generate responses and fine tune the smaller model.