As cloud computing evolves, Amazon's AWS looks for its role in generative AI

Bob O'Donnell

Posts: 88   +1
Staff member

One of the intriguing aspects of the generative AI phenomenon is the vast range of vendors that are offering solutions to leverage the new technology. Companies eager to deploy GenAI face a complex landscape filled with foundation model suppliers, AI platform companies, data management vendors, and model customization tool providers.

Surprisingly, the big cloud computing companies – which have dominated the IT landscape for the last decade – haven't played as central a role as many initially expected. At least, not yet.

But there are signs that situation could be changing. Google recently held their Cloud Next event where they unveiled a wide range of new AI integrations for Google Workspace, a GenAI tool for video creation and editing, and other enhancements aided by its Gemini Pro 1.5 large language model.

Now it's Amazon AWS taking the wraps off a host of new features and improvements for its Bedrock GenAI managed service, which are designed to make the process of selecting and deploying the right tools for GenAI applications much easier. Amazon is adding the ability to import customized foundation models into the service and then allow companies to leverage the capabilities of Bedrock across those custom models.

For example, companies that have trained an open-source model like Llama or Mistral with their own data, possibly using Amazon's SageMaker, can now integrate that customized model along with the existing standardized models within Bedrock. This integration allows the use of a single API to build applications that utilize both customized and existing Bedrock model options, including those from AI21 Labs, Anthropic, Cohere, Meta, Stability AI, and Amazon's Titan models.

Amazon also introduced version 2 of their Titan Text Embeddings model, which has been specifically optimized for RAG (Retrieval Augmented Generation) applications and announced the general availability of its Titan Image Generator model.

The ability to import custom models into Bedrock supports the integration of RAG functionalities, facilitating continuous fine-tuning of models with new data. Bedrock's serverless infrastructure also supports scalable performance across AWS instances, which aids in managing real-time demands. Furthermore, Bedrock includes tools for developing AI-powered agents capable of performing multi-step tasks. Agents are currently one of the hottest discussion topics in GenAI, so these kinds of capabilities are bound to be of interest to those organizations that want to stay on the cutting edge.

Additionally, Amazon has introduced new Guardrails for Bedrock, which add enhanced filtering features to prevent the creation and dissemination of inappropriate content and sensitive information. While existing models already incorporate basic content filtering, these new Guardrails offer an extra layer of customizable protection.

Amazon's Model Evaluation tool within Bedrock, now generally available, assists organizations in selecting the most suitable foundation model for their specific needs. This tool compares various models based on accuracy, robustness, and allows customization to evaluate how different models perform with user-specific data and prompts.

As organizations navigate the early stages of GenAI deployments, many are realizing the importance of locating their GenAI software and services near their data sources. With a significant amount of data hosted on AWS, the new features in Bedrock could prove particularly attractive to companies looking to enhance their GenAI capabilities.

We may also witness the emergence of multi-platform GenAI deployments. Just as companies have found benefits in using multiple cloud providers, they are likely to adopt a similar strategy with GenAI platforms, using different platforms for various applications.

The race is still on, but it's clear that all the major cloud computing providers want to be (and will be) important entrants in the GenAI landscape as well.

Bob O'Donnell is the founder and chief analyst of TECHnalysis Research, LLC a technology consulting firm that provides strategic consulting and market research services to the technology industry and professional financial community. You can follow him on Twitter

Permalink to story:

 
Back