AI in sports: How AI and Amazon Bedrock are changing the game (SMB101)

Summary

The Story of Infinite Athlete

  • In 2018, two significant events occurred:
    • The landmark "Attention is All You Need" paper was published, leading to the rise of GPT-1 and other transformer-based language models.
    • The Alliance of American Football (AAF) was formed, providing the opportunity to build a national sports league's technology stack from the ground up.
  • The sports industry was largely dependent on traditional methods, such as yellow legal pads, for data analysis and insights.
  • Infinite Athlete aims to revolutionize the sports industry by:
    • Integrating data from various sources to create comprehensive, high-quality datasets.
    • Leveraging these datasets and advanced AI models to drive research and insights that directly impact player health, safety, and performance.

Bedrock: Powering Generative AI Applications

  • Amazon Bedrock is a fully managed service that provides access to advanced foundation models from leading AI innovators.
  • Bedrock allows for customization of these models with your data, using techniques like fine-tuning and prompting.
  • It is integrated with other AWS services, enabling easy integration with your existing application landscape.
  • Bedrock offers security features, guard rails for responsible AI, and options for model distillation and prompt routing.

A Practical Use Case: Bridging the Gap between Users and Data

  • Infinite Athlete uses Bedrock to build solutions that allow users to query data using natural language.
  • The approach involves:
    • Passing the user's input and relevant context to the Bedrock model.
    • Asking the model to generate a database query that fulfills the user's request.
    • Executing the query against the database and returning the results to the user.
  • This approach has several advantages:
    • Allows for exact matches on complex criteria.
    • Enables returning large datasets that may not fit in the model's context window.
    • Requires only a single model invocation, making the solution fast.
    • Ensures the model never directly touches the data, addressing security concerns.
    • Allows for caching of model outputs, improving performance and cost-efficiency.

Ensuring Successful Generative AI Journeys

  • The key to successful generative AI implementation is identifying and fleshing out the right business use cases.
  • AWS solution architects, TAMs, CSMs, and account teams can help customers in this process, from building the business justification to visualizing the desired outcomes.
  • AWS also has various channel partners and funding mechanisms to support customers in their generative AI journeys.

Your Digital Journey deserves a great story.

Build one with us.

Cookies Icon

These cookies are used to collect information about how you interact with this website and allow us to remember you. We use this information to improve and customize your browsing experience, as well as for analytics.

If you decline, your information won’t be tracked when you visit this website. A single cookie will be used in your browser to remember your preference.

Talk to us