AWS’ New AI Security ‘A Huge Confidence Boost’ For GenAI Customer Adoption

Here’s why AWS’ new Guardrails for Amazon Bedrock cloud security offering will drive enterprises to deploy more generative AI solutions.

ARTICLE TITLE HERE

With some businesses hesitant to deploy generative AI due to security concerns, AWS’ new launch of Guardrails for Amazon Bedrock will give customers the confidence needed to dive deeper into GenAI.

“Everybody’s racing for a new GenAI platform to launch,” said Scott White, chief operating and revenue officer for AWS partner DoiT International. “But what companies really need in order to adopt GenAI is guardrails and governance.”

“At this point, there’s been very little work on, ‘How do I go deploy this in an enterprise wide-fashion with some confidence that my team is going to do what I intended them to do with it. And have the right guardrails and checkpoints and governance in place,’” said White, adding that his company’s AWS business spiked by a whopping 400 percent this year. “So this GenAI Guardrails [for Amazon Bedrock] announcement is massive. It’s going to be a huge confidence boost to customers to go to deploy GenAI.”

At AWS re:Invent 2023 in Las Vegas this week, AWS unveiled Guardrails for Amazon Bedrock in a move to implement safeguards customized to a business’s application requirements.

AWS CEO: Guardrails Provides ‘Consistent Level of Protection’

The new AI cloud security offering evaluates user inputs and foundational models (FM) responses based on use case-specific policies, and provides an additional layer of safeguards regardless of the underlying FM. Customers can create multiple guardrails, each configured with a different combination of controls, and use these guardrails across different applications and use cases.

Amazon Bedrock is AWS’ generative AI platform to enable customers to easily build and scale GenAI applications with various foundation models from AWS and third parties such as Anthropic and Cohere.

AWS CEO Adam Selipsky said his $92 billion Seattle-based global cloud market share leader has been working diligently on providing responsible AI solutions.

“We’ve been actively participating, and a lot of groups have come together, to discuss responsible AI issues. Earlier this year, I joined [U.S.] President [Joe] Biden and other leaders. We’ve made a series of voluntary commitments promoting safe, secure and transparent development of AI technology,” said Selipsky during his keynote at AWS re:Invent 2023. “We’ve been working hard here, and today, we’re announcing Guardrails for Amazon Bedrock.”

“Now you have a consistent level of protection across all of your GenAI development activities,” said AWS’ CEO. “So we’re approaching the whole concept of generative AI in a fundamentally different way because we understand what it takes to reinvent how you’re going to build with this technology.”

Guardrails for Amazon Bedrock can also be integrated with Agents for Bedrock to build generative AI applications aligned with a customers’ responsible AI policies. The offer also integrates with Amazon CloudWatch so users can monitor and analyze user inputs and FM responses that violate policies defined in the guardrails.

AWS Is ‘A Step Ahead’ Of Others

Santa Clara, Calif.-based DoiT International is an all-star cloud services provider and an AWS Premier Tier Servies partner with more than 300 certifications.

This month, the global AWS partner signed a significant strategic collaboration agreement with AWS aimed at generating $5 billion in business by driving AWS cloud consumption for born-in-the-cloud to large enterprise customers.

Guardrails for Amazon Bedrock will help drive AWS momentum as it will be a huge confidence builder and booster for companies to implement generative AI, said Tracie Stamm, head of product marketing and content for DoiT. Stamm said AWS is a “step ahead” of some of the other GenAI market players due to the cloud giant’s massive scale and enterprise focus with Guardrails.

“Just the scale of AWS means they care about being able to bring everybody along on this curve,” said Stamm. “Smaller companies that don’t have the same security sensitivities, they’re able to be early adopters. But you see these larger enterprises—like Feds, SLED, financial services—are much tighter in their security requirements. You can’t get that business if you don’t provide the right controls, governance and oversight.”

Guardrails for Amazon Bedrock is currently available in limited preview. During preview, Guardrails can be applied to all LLMs available on Amazon Bedrock including Amazon Titan Text, Anthropic Claude, Meta Llama 2, AI21 Jurassic, and Cohere Command.