Skip to content

Contact sales

By filling out this form and clicking submit, you acknowledge our privacy policy.

AWS re:Invent 2023: Highlights from CEO Adam Selipsky’s keynote

From fresh cloud transformation insights to innovative artificial intelligence announcements, here’s what went down on Tuesday at AWS re:Invent 2023.

Nov 28, 2023 • 4 Minute Read

Please set an alt value for this image...
  • Cloud
  • IT Ops
  • Software Development
  • Engineering Leadership
  • Data
  • Security
  • Business
  • News
  • AI & Machine Learning
  • Learning & Development
  • AWS

The second full day of AWS re:Invent 2023 kicked off with a keynote from the CEO of Amazon Web Services, Adam Selipsky. He shared his perspective on cloud transformation and provided an exclusive first look at AWS’s innovative artificial intelligence, machine learning, and infrastructure advancements.

In this post, we break down the highlights and explain what you need to know to fully leverage cloud and AI technology. Now sure what to watch during the rest of the event? Check out our guide to the top sessions and speakers at this year’s AWS re:Invent.

Table of contents

Investing in the Generative AI Stack

“GenAI is the next step in artificial intelligence,” said Adam. He explained how AWS is investing in AI across three levels of what he called the Generative AI Stack. 

  1. Infrastructure for training and inference

  2. Tools to build with large language models (LLMs) and other foundational models (FMs)

  3. Applications that leverage FMs

Each new or enhanced product announcement fell into one of these levels. Let’s take a look at a few of the most exciting ones.

Enhancing cloud storage with Amazon S3 Express Zone One

Adam started off by unveiling Amazon S3 Express One Zone, a new Amazon S3 storage class. It’s built to deliver high-performance and low-latency object storage for organizations’ most frequently accessed data.

It boasts single-digit millisecond latency and data access speeds up to 10 times faster and request costs up to 50% lower than Amazon S3 Standard. The bottom line? It’s faster and more cost-efficient.

Adam shared the example of Pinterest: Using S3 Express Zone One, they saw a 40% reduction in total costs that enabled faster personalization and an improved user experience.

Amazon Q: The generative-AI-powered assistant

One of the most exciting announcements was Amazon Q. What is Amazon Q? It’s a new generative-AI-powered assistant for work that provides quick, accurate, and relevant answers to pressing business questions tailored to each employee’s role and permissions. It’s built to be secure and will be a huge time saver for cloud practitioners (and non-technical employees, too!).

You can use it to build applications on AWS, research best practices, respond to incidents, get help coding new features, and more. It can even be configured to create Jira tickets and create Slack notifications.

Amazon DataZone AI recommendations for simplified data cataloging

“In generative AI, your data is your differentiator,” said Adam. The new Amazon DataZone AI recommendations uses generative AI so you can add business context to your data catalog in a few clicks and get recommendations on how to use the data for faster insights and action.

New zero-ETL integrations enable faster data analysis

Adam announced four new zero-ETL integrations that automatically connect data from the source to the destination and eliminate the burden of extract, transform, and load (ETL) jobs.

  1. Amazon Aurora PostgreSQL zero-ETL integration with Amazon Redshift

  2. Amazon RDS for MySQL zero-ETL integration with Amazon Redshift

  3. Amazon DynamoDB zero-ETL integration with Amazon Redshift

  4. Amazon DynamoDB zero-ETL integration with Amazon OpenSearch Service

AWS Graviton4 and AWS Trainium2 power high performance AI and ML training

As organizations build AI and ML models, they’ll need the processing power to go along with it. But, Adam explained, they won’t need the same amount all the time. Fluctuating demands call for short-term clustered capacity.

In response, AWS created Graviton4 and Trainium2. Together, they’ll enable high-performance, cost-effective generative AI and ML training. According to AWS, “AWS Graviton4 is the most powerful and energy-efficient AWS processor to date for a broad range of cloud workloads. AWS Trainium2 will power the highest performance compute on AWS for training foundation models faster and at a lower cost, while using less energy.”

New Amazon Bedrock capabilities

Adam also shared a wide range of new Amazon Bedrock capabilities.

Learn how to get started with Amazon Bedrock.

Guardrails for Amazon Bedrock: Prioritizing responsible AI

Guardrails for Amazon Bedrock allow you to safeguard your AI applications by applying guardrails to your large language models (LLMs). This helps ensure responsible, ethical AI use that follows your organization’s policies and best practices.

Agents for Amazon Bedrock: Execute multi-step tasks with AI technology

Agents for Amazon Bedrock is now generally available. Agents uses generative AI to execute multi-step tasks like answering customer questions or taking a sales order.

Enhanced Amazon Bedrock customization capabilities

As Adam noted, “You need generative AI that understands your business, your customers, your products, [and] your operations.”

That’s only possible when you can use your own data to customize the models. That’s why AWS has introduced fine tuning, retrieval augmented generation (RAG), and continued pre-training that will allow you to tailor generative AI to your business.

Expanding Salesforce, Nvidia, and Anthropic partnerships

Along with the exciting product and service announcements, Adam also explained how AWS is expanding partnerships with Salesforce, Nvidia, and Anthropic.

Salesforce will now support Amazon Bedrock “to power AI-driven apps and workflows in Salesforce.”

In addition, Nvidia founder and CEO Jensen Huang took the stage to explain how AWS and Nvidia are working together to build a cloud AI supercomputer and reduce the training time of the largest language models.

Dario Amodei, CEO and co-founder of Anthropic, also spoke about the expanded partnership with Amazon, how Claude 2.1 is targeted for enterprises, and the need to evaluate and measure models.

Explore more AWS resources

Adam ended the keynote with a look at Project Kuiper, Amazon’s initiative to provide fast, reliable internet to communities around the world. 

Throughout it all, Adam made sure to emphasize how these new products and services will help organizations deliver greater customer value—and the cloud and AI skills their teams need to keep up. 

Want to keep your cloud momentum going? Start a free trial and check out our hands-on labs for AWS: