Generative AI on AWS
Tools for building with
generative AI
© 2023, Amazon Web Services, Inc. or its affiliates. All rights reserved. © 2023, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Introduction
Machine learning: Amazon’s heritage, ethos,
and future
Agenda
Making generative AI straightforward and
accessible
How can you get started today?
© 2023, Amazon Web Services, Inc. or its affiliates. All rights reserved. 2
Machine learning (ML) is at an inflection point
Key drivers: Compute capacity increase | Data growth | Model sophistication
© 2023, Amazon Web Services, Inc. or its affiliates. All rights reserved. 3
ML innovation is in the Amazon DNA
4,000 products 1.6M packages Billions of Alexa First Prime Air delivery
per minute sold every day interactions each week on December 7, 2016
on Amazon.com
© 2023, Amazon Web Services, Inc. or its affiliates. All rights reserved. 4
More than 100,000 customers use AWS for ML
© 2023, Amazon Web Services, Inc. or its affiliates. All rights reserved. 5
Our focus has been on democratizing ML
The broadest and AWS DeepRacer Training and AWS AI/ML Machine Learning
deepest AI/ML League certification Scholarship University
capabilities
© 2023, Amazon Web Services, Inc. or its affiliates. All rights reserved. 6
Question: What is generative artificial intelligence (AI)?
• Creates new content and ideas, including conversations,
stories, images, videos, and music
• Powered by large models that are pretrained on vast corpora
of data and commonly referred to as foundation models (FMs)
BLOOM
© 2023, Amazon Web Services, Inc. or its affiliates. All rights reserved. 7
Why foundation models?
Tasks Tasks
… … Text generation Text generation
… … Summarization Summarization
Labeled ML Information Unlabeled Information
FM
data models extraction data extraction
… … Q&A Q&A
… … Chatbot Chatbot
Train Deploy Pretrain Adapt
© 2023, Amazon Web Services, Inc. or its affiliates. All rights reserved. 8
Flexibility
Secure
customization
The most cost-
Why AWS for effective
generative AI? infrastructure
The easiest way
to build with FMs
Generative AI-
powered solutions
© 2023, Amazon Web Services, Inc. or its affiliates. All rights reserved. 9
Flexibility Choose from a wide
selection of FMs built by
AI21 Labs, Anthropic,
Secure
Stability AI, and Amazon to
customization
find the right model for your
use case.
The most cost-
Why AWS for effective
generative AI? infrastructure
The easiest way
to build with FMs
Generative AI-
powered solutions
© 2023, Amazon Web Services, Inc. or its affiliates. All rights reserved. 10
Flexibility
Secure Customize FMs for your
customization business with just a few
labeled examples. Since all
data is encrypted and does
The most cost-
Why AWS for effective
not leave your Amazon
Virtual Private Cloud (VPC),
generative AI? infrastructure you can trust that your data
will remain private and
The easiest way confidential.
to build with FMs
Generative AI-
powered solutions
© 2023, Amazon Web Services, Inc. or its affiliates. All rights reserved. 11
Flexibility
Secure
customization
The most cost-
Why AWS for effective
Get the best price
generative AI? infrastructure
performance for generative
AI with infrastructure
powered by AWS-designed
The easiest way ML chips and NVIDIA GPUs.
to build with FMs Cost-effectively scale
infrastructure to train and
run FMs containing hundreds
Generative AI- of billions of parameters.
powered solutions
© 2023, Amazon Web Services, Inc. or its affiliates. All rights reserved. 12
Flexibility
Secure
customization
The most cost-
Why AWS for effective Quickly integrate and deploy
generative AI? infrastructure FMs into your applications
and workloads running on
AWS using familiar controls
The easiest way
and integrations with the
to build with FMs
depth and breadth of AWS
capabilities and services like
Generative AI- Amazon SageMaker and
powered solutions Amazon S3.
© 2023, Amazon Web Services, Inc. or its affiliates. All rights reserved. 13
Flexibility
Secure
customization
With generative AI built in,
The most cost- services such as Amazon
Why AWS for effective CodeWhisperer, an AI coding
generative AI? infrastructure companion, can help you
improve productivity. In
addition, you can deploy
The easiest way common generative AI use
to build with FMs cases like call summarization
and question answering using
AWS sample solutions that
Generative AI-
combine AWS AI services with
powered solutions
leading FMs.
© 2023, Amazon Web Services, Inc. or its affiliates. All rights reserved. 14
Building with generative AI on AWS
© 2023, Amazon Web Services, Inc. or its affiliates. All rights reserved. 15
NEW
Amazon Bedrock
The easiest way to build and
scale generative AI
applications with FMs
© 2023, Amazon Web Services, Inc. or its affiliates. All rights reserved. 16
Amazon Bedrock key benefits
Accelerate Choose FMs from Privately customize Enhance your data Use AWS tools and
development of AI21 Labs, Anthropic, FMs using your protection using capabilities that you
generative AI Stability AI, and organization’s data comprehensive AWS are familiar with to
applications using Amazon to find the security capabilities deploy scalable,
FMs through an API, right FM for your use reliable, and secure
without managing case generative AI
infrastructure applications
© 2023, Amazon Web Services, Inc. or its affiliates. All rights reserved. 17
Bedrock supports a wide range of foundation models
FMs from Amazon FMs from AI21 Labs, Anthropic, and Stability AI
Titan Text Titan Jurassic-2 Claude Stable
Embeddings Diffusion
© 2023, Amazon Web Services, Inc. or its affiliates. All rights reserved. 18
Amazon Titan
INNOVATE RESPONSIBLY WITH HIGH-
PERFORMING FMs FROM AMAZON
Benefits
• Built with 20+ years of Amazon
ML experience
• Automate language tasks such as
summarization and text generation with
Amazon Titan Text FM
Titan Text Titan Embeddings
• Enhance search accuracy and improve
focused on for enterprise tasks
personalized recommendations with
NLP tasks such as search and
Amazon Titan Embeddings FM
personalization
• Support responsible use of AI by reducing
inappropriate or harmful content
© 2023, Amazon Web Services, Inc. or its affiliates. All rights reserved. 19
Foundation models from top AI startups
Jurassic-2 Claude Stable Diffusion
Multilingual LLMs for text LLM for conversational and text Generation of unique, realistic,
generation in Spanish, French, processing tasks high-quality images, art, logos,
German, Portuguese, Italian, and and designs
Dutch
© 2023, Amazon Web Services, Inc. or its affiliates. All rights reserved. 20
Privately customize foundation models using your
organization’s data
Fine-tune
PURPOSE Maximizing accuracy for specific tasks
DATA NEED Small number of labeled examples
© 2023, Amazon Web Services, Inc. or its affiliates. All rights reserved. 21
Amazon Bedrock brings quality, scalability,
and performance to Coda AI
As longtime happy AWS customers, we’re excited about how
Amazon Bedrock can bring quality, scalability, and performance to
Coda AI. We also appreciate that we can use Amazon Titan models
through a simple API without provisioning or managing
infrastructure. Since all our data is already on AWS, we are able to
quickly incorporate generative AI using Bedrock, with all the
security and privacy we need to protect our data built in. With
over tens of thousands of teams running on Coda, including large
teams like Uber, the New York Times, and Square, reliability and
scalability are really important.
Shishir Mehrotra
Co-Founder and CEO
© 2023, Amazon Web Services, Inc. or its affiliates. All rights reserved. 22
decrease in computing costs
Accelerate your generative AI initiatives with the
global partner community
At Deloitte, we’re advancing our generative AI capabilities to help
clients harness its power. As part of that effort, the organization is
leveraging leading offerings like Amazon Bedrock to advance such
capabilities with our alliance relationships. With Bedrock, we can
provide clients with a cost-effective serverless API for AWS
customers to build generative AI applications. We are eager to
help clients deliver better, faster AI results to the mutual
customers of AWS and Deloitte.
Nishita Henry
Amazon/AWS Alliance Global Chief Commercial Officer, Deloitte Consulting LLP
© 2023, Amazon Web Services, Inc. or its affiliates. All rights reserved. 23
decrease in computing costs
Accelerate your generative AI initiatives with the
global partner community
Slalom’s deep partnership with AWS thrives on a shared
commitment to obsessively provide our customers with
exceptional value. We are proud to support AWS’s innovative
generative AI toolkits, which are purpose-built to help enterprise
customers create real-world benefits in this fast-changing
environment.
Mukesh Kumar
Global GM of Data and Technology, Slalom Consulting LLC
© 2023, Amazon Web Services, Inc. or its affiliates. All rights reserved. 24
decrease in computing costs
NEW
Amazon EC2 Trn1n and
Amazon EC2 Inf2
Purpose-built ML accelerators for
the best price performance for
training and inference in the cloud
© 2023, Amazon Web Services, Inc. or its affiliates. All rights reserved. 25
Purpose-built accelerators for generative AI
AWS Inferentia AWS Trainium AWS Inferentia2
Lowest cost per inference The most cost-efficient, high- High performance at the
in the cloud for running performance training of lowest cost per inference for
deep learning (DL) models LLMs and diffusion models LLMs and diffusion models
Up to 70% lower Up to 50% savings Up to 40% better
cost per inference on training costs price performance
than comparable over comparable than comparable
Amazon EC2 instances Amazon EC2 instances Amazon EC2 instances
© 2023, Amazon Web Services, Inc. or its affiliates. All rights reserved. 26
Runway delivers a better creator experience with AWS
Inferentia2
At Runway, our suite of AI Magic Tools enables our users to
generate and edit content like never before. With Amazon
EC2 Inf2 instances, we’re able to run some of our models
with up to 2x higher throughput. This enables us to
introduce more features, deploy more complex models,
and ultimately deliver a better experience for the millions
of creators using Runway.
Cristobal Valenzuela
CEO
© 2023, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Finch Computing saved costs with AWS
Inferentia
CHALLENGE SOLUTION
Needed cost-effective ways Finch Computing uses AWS Inferentia for FM inference for
to use GPUs to support FM many use cases, including language translation, text
inference tasks summarization, and headline generation
OUTCOME
ü 80% reduction in costs
ü Expanded support for three additional languages
ü Accelerated time to market
© 2023, Amazon Web Services, Inc. or its affiliates. All rights reserved. 28
4/26/23
decrease in computing costs
NEW
Amazon CodeWhisperer
Build applications faster and
more securely with an AI coding
companion
© 2023, Amazon Web Services, Inc. or its affiliates. All rights reserved. 29
Amazon CodeWhisperer: Now generally available and
free to use for individual developers!
Generate code Scan code for hard-to- Flag code that resembles
suggestions in real time find vulnerabilities open-source training
data or filter by default
During preview Amazon ran a productivity challenge, and participants who used
Amazon CodeWhisperer were 27% more likely to complete tasks successfully and did
so an average of 57% faster than those who did not use CodeWhisperer.
© 2023, Amazon Web Services, Inc. or its affiliates. All rights reserved. 30
Accenture improves developer productivity with
Amazon CodeWhisperer
Accenture is using Amazon CodeWhisperer to accelerate coding as
part of our software engineering best practices initiative in our
Velocity platform. The Velocity team was looking for ways to
improve developer productivity. After searching for multiple
options, we came across Amazon CodeWhisperer, which reduced
our development efforts by 30%. We are now focusing more on
improving security, quality, and performance.
Balakrishnan Viswanathan
Tech Architecture Sr. Manager, Accenture
© 2023, Amazon Web Services, Inc. or its affiliates. All rights reserved. 31
Start your generative AI journey today
Watch Werner
Check out the Read Vogels, AWS VP
generative AI announcement
and CTO, explain
webpage blog post
generative AI
© 2023, Amazon Web Services, Inc. or its affiliates. All rights reserved. 32
Thank you!
© 2023, Amazon Web Services, Inc. or its affiliates. All rights reserved. © 2023, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Amazon Web Services, AWS, the Powered by AWS logo, and all AWS service
names used in this slide deck are trademarks of Amazon.com, Inc. or its affiliates.
