Category: Foundation models

Many organizations validate their generative AI–based workloads on their emotional reaction to a foundation model’s output instead of real testing criteria. In this post, we show you how you can move away from vibe testing and effectively evaluate your HAQM Bedrock workloads for production.

Today, generative artificial intelligence (AI) is at the core of many of the decisions and technologies businesses across the globe are implementing. The linchpin to ensuring that generative AI is effective depends on the foundation models (FMs) storing the data. To make sure that those FMs are pulling the correct data, HAQM Web Services (AWS) has developed a solution through HAQM Bedrock to generate SQL to answer user questions about the data. The solution was developed to support the Federal Institute of São Paulo (IFSP) and has optimized their decision-making.

HAQM Bedrock is now generally available in the HAQM Web Services (AWS) Top Secret cloud. With authority to operate in the AWS Top Secret Cloud, HAQM Bedrock is providing U.S. Government customers with secure access to foundation models (FMs) and generative artificial intelligence (AI) capabilities within AWS’s classified environment.