Why AWS Partners Are Excited About the Latest Innovations in Generative AI on AWS

By Priya Arora, Global Head of Generative AI Center of Excellence – AWS Partner Organization

Enterprises around the world are exploring the possibilities and potential of generative AI, and realizing the inherent complexity associated with leveraging it to address business challenges. This is why Amazon Web Services (AWS) made the commitment to simplify and democratize generative AI.

Our latest offerings, many of which were announced during the 2023 New York Summit, are a testament to this commitment. Thanks to these innovations and AWS’s long-standing history of artificial intelligence (AI) innovation, we have seen how you, our trusted AWS Partners, have identified AWS as a premier choice in generative AI. We’ve also seen how pleased you are with the many ways these innovations will support your customers.

For example, IBM Consulting and AWS are working together to deliver generative AI capabilities to help France-based Bouygues Telecom up-level their customer relationships and take advantage of millions of previous conversations with customers. With the latest AI technologies, Bougyes Telecom can now streamline call summarization, making it easier to turn previous customer calls into valuable business intelligence.

“We know that clients today are looking for the technical support and AI expertise needed to help them build a strategy and develop generative AI use cases that will drive business value and transformation,” says Manish Goyal, Sr. Partner, Global AI and Analytics Leader at IBM Consulting. “Our expertise in helping clients innovate on AWS, combined with AWS’s recently announced offerings, will help our joint clients to scale generative AI rapidly and responsibly.”

AWS has also seen partners like Persistent use Amazon Bedrock to provide generative AI tools to customers like Blast Motion so they can continue to revolutionize the digital transformation of sports.

Simultaneously, other partners like AIShield – a Bosch startup are enhancing AWS AI services with integrations that ensure customers enjoy a safe and secure experience as they use generative AI.

Seeing the cross-industry applications our customers are able to support, it’s exciting to think about the scale at which generative AI can make a positive impact. During our recent New York Summit, we announced seven new updates that will help AWS Partners build generative AI applications across industries including enterprise tech, healthcare, retail, and software development.

Partners Recognize the Business Value in Generative AI on AWS

Each AWS announcement highlights a product or service that’s designed to make generative AI more accessible, easier to use, and more valuable to business success.

AWS recognizes that if our partners effectively leverage generative AI offerings, they can increase how they provide an excellent client experience and/or empower their teams toward business goals.

Here’s what AWS Partners are saying about each offering:

Agents for Amazon Bedrock Increases Efficiency and Accuracy

Agents for Amazon Bedrock allows developers to use foundational models (FMs) to complete actions based on company data and user input.

“Agents for Amazon Bedrock is the killer feature,” says Randall Hunt, VP of Cloud at Caylent. “It has applications in every segment and vertical. Agents provides a natural language interface over APIs and removes the need for millions of lines of error-prone parsing code written over the last 20 years. We already have customers interested in replacing their email parsing code with Agents, their automated IVR flows with Agents, and even enabling Agents to take actions on their AWS infrastructure in response to errors in CloudWatch logs.”

New AWS Foundation Models Increase Flexibility and Choice

We also announced two new updates to AWS FM offerings. First, Stability AI is making available the second iteration of its text-to-image model available to developers. Also, Amazon Bedrock now includes Cohere as its latest FM provider. Cohere joins AWS’s best-in-class models such as AI21 Labs, Anthropic, Stability AI, and Amazon, which gives our partners more choice and flexibility.

“Amazon Bedrock provides enterprises access to Cohere’s multi-lingual embeddings equipped to read and understand over 100 languages and text generation command models for summarization, copywriting, dialogue, extraction, and question answering,” says Martin Kon, President and COO at Cohere. “This is a huge step forward in adding flexibility for companies looking to harness the potential of generative AI. Bedrock customers can now leverage our models to build and scale generative AI applications on-demand in their AWS environment.”

Amazon CodeWhisperer Integration with Amazon Glue Studio Notebooks

When using Amazon Glue Studio to power extract, transform, and load (ETL) workflows, developers can now use Amazon CodeWhisperer to get suggestions on the best ways to streamline their data transfer and data management.

AWS JAM event attendees were able to test this new developer experience with CodeWhisperer. During the CodeWhisperer preview, AWS ran a productivity challenge and participants who used the service were 27% more likely to complete tasks successfully and did so an average of 57% faster than those who did not use CodeWhisperer.

“Through the AWS JAM event, the Wipro AWS developer community has witnessed the power of CodeWhisperer and how it can make the day in the life of a developer easier and more productive,” says Ramesh Pai, VP, AWS Business Unit Head at Wipro FullStride Cloud. “During the event, developer teams were able to easily solve the challenges from the AWS team—in just a limited time—using CodeWhisperer. After the event, 91% of participants said the CodeWhisperer JAM event helped them learn something new.”

Amazon EC2 P5 Instances Increase Cost Efficiencies

With Amazon EC2 P5 instances powered by the latest NVIDIA H100 Tensor Core GPUs, AWS Partners can accelerate their time to solution by up to 6x compared to previous-generation GPU-based EC2 instances, and reduce the cost to train machine learning (ML) models by as much as 40%.

“AWS and NVIDIA’s GPUs can help provide our clients performance with endless possibilities,” says Nitin Mittal, Principal, US Generative AI Innovation Leader at Deloitte Consulting LLP. “Deloitte is at the forefront of the pioneering efforts to transform the world of data processing, analytics, and AI. Amazon EC2 P5 instances allow us to combine Deloitte experience with AWS Cloud and Nvidia GPUs to generate meaningful insights, co-innovate with our clients, and develop novel solutions that get to the heart of business challenges.”

Amazon HealthScribe Enhances Productivity within the Healthcare Segment

Amazon HealthScribe allows clinicians to automatically generate notes by analyzing patient-clinician conversations in clinical applications. Our partners at Automation Anywhere, who have expertise in streamlining the clinical note-taking process, are looking forward to how HealthScribe will alleviate some of the note-taking pressure on clinical staff.

“Synthesizing clinical notes is tedious, time-consuming, and error-prone for busy physicians and nurses,” says Dr. Yan Chow, a physician technologist and Automation Anywhere’s Global Healthcare Leader. “Creating patient after-visit summaries using the HIPAA-eligible AWS HealthScribe represents a groundbreaking approach that combines the powerful capabilities of intelligent automation with generative AI. This saves clinicians hours of work, improves document accuracy, and boosts patient satisfaction.”

Amazon QuickSight Accelerates Time to Insights with Generative BI Capabilities

Amazon QuickSight simplifies the path from data to insights by leveraging the power of generative AI in the business intelligence (BI) lifecycle. The financial services industry is one area where the generative BI capabilities in QuickSight has the potential to have a major impact.

“At Capgemini, we understand just how generative AI has the potential to revolutionize financial services,” says Shashi Gupta, VP and Global Head of AWS COE for Financial Services at Capgemini. “Financial services institutions analyze petabytes of data from various sources such as portfolios, actuarial, risk management systems, and historical stock prices. The generative BI capabilities in QuickSight can help us build intuitive and robust natural language experiences to accelerate the path to meaningful insights.”

Vector Engine for Amazon OpenSearch Serverless Expedites Processing Time

With the high-performing Vector Engine for Amazon OpenSearch Serverless, developers can build intuitive ML-augmented experiences—all without the need to manage vector database infrastructure.

“We are particularly thrilled about the launch of the Vector Engine for Amazon OpenSearch Serverless,” says Charles Burden CEO at TensorIoT. “This is a game-changer in the field of generative AI, particularly for our customers in the financial, legal, and healthcare sectors. Vector Engine allows us to craft advanced, highly secure, and compliant generative AI solutions that drive business transformation by refining and accelerating processes.”

Getting Started with New Generative AI Offerings on AWS

We’re excited that so many AWS Partners see the potential in how each new generative AI offering can enhance business operations and customer relations. As customers also recognize this potential, they are looking for support in how they implement generative AI.

The best preparation for you to provide this level of support to customers is through specializing in use cases, familiarizing with generative AI tools on AWS, building relevant solutions, and engaging with customers on how the right solutions can both increase productivity and be cost effective.

Together, we can focus on four clear actions: Specialize, Familiarize, Build, and Engage Customers.

#1 – Specialize

We’re currently seeing momentum around certain horizontal use cases for generative AI tools. If you’re able to apply an industry lens to each of these use cases, it will help customers with quicker adoption of generative AI:

• Enhance customer experiences with chatbots, virtual assistants, and more personalized content.
• Increase employees’ productivity with conversational search, code generation, or text summarization.
• Accelerate content creation with relevant suggestions and technology that turns text to images, artwork, or stories.
• Improve operations through faster document processing and more efficient quality control processes.

Figure 1 – Examples of horizontal generative AI use cases.

Since business use cases span industries, it’s likely customers will approach AWS Partners with issues within these four categories. As an expert in how generative AI can alleviate specific issues in each category, you’ll be equipped to discuss unique challenges with customers and provide solutions that power every use case.

#2 – Familiarize

After you grasp specific use cases, it’s important to also understand the right generative AI tools for your customers. For example, AWS offers best-in-class FMs such as AI21 Labs, Anthropic, Cohere, Stability AI, and Amazon. With several models to choose from, developers should familiarize themselves with each model and see how they relate to many of the business problems within your industry.

It may help to ask questions like:

• What is the modality of the use case (text, image, vector/embedding)?
• What language, task, or latency is acceptable for the use case? (This will determine model applicability and size.)
• What degree of customization is required to satisfy application requirements (zero or few shot prompting vs. fine-tuning)?

By getting familiar with each model, you’ll be able to answer these questions correctly and offer generative AI solutions that directly meet customers’ needs.

#3 – Build

Many AWS Partners have already developed solutions that are helping customers realize the potential of generative AI. AWS Partners have built over 60 demos covering more than 50 distinct use cases.

However, new use cases are arising daily, and for many customers the application of AI is still just a possibility. We encourage each partner to select a relevant use case and build a demo to create a proof of concept (PoC) to address a current customer problem.

Building these demos and applying them to real use cases will help move customers past the art of what’s possible with AI. As you build out each demo, the AWS Partner Network (APN) will better understand partner capabilities to share with customers and AWS Sellers.

#4 – Engage

Building demos is just the start. As you engage your customers on how they can leverage generative AI, use each demo to create playground experiences so they can test the real-world applications of the technology. This is especially important in front of senior leadership who will have questions about justifying spend and resources required for implementation.

Figure 2 – AWS VP Swami Sivasubramanian speaking on stage at 2023 New York Summit.

Learn More About AWS and Generative AI

At AWS, we feel it’s necessary to provide the tools to start the generative AI journey and walk with our partners on that journey. This is why we’re excited about our generative AI digital training that’s now available, courtesy of the AWS Generative AI Center for Excellence. Partners will have a chance to complete this training and, upon competition, receive a badge that affirms their expertise in AI.

We also encourage you to take part in our seven free and low-cost generative AI courses. These comprehensive, on-demand courses allow you to up-level your generative AI skills as you seek to serve customers’ AI needs.

AWS Partners can learn more by registering for PartnerCasts on Amazon Bedrock business and technical training on October 9, and technical training on October 10.