Have you convinced your boss yet? Groups get the best deals 🎟️ Buy now before price increase →

This article was published on January 14, 2022

Why is the NFL crowd-sourcing feckless AI solutions to its concussion problem?

What's the point?


Why is the NFL crowd-sourcing feckless AI solutions to its concussion problem?

The NFL and Amazon Web Services (AWS) today announced the results of their second annual artificial intelligence competition. In total, five teams split the $100K prize. And I can’t think of any reason why you should care.

Up front: According to the NFL, the contest is supposed to help solve its injury problems using machine learning.

Here’s what the league had to say in today’s press release:

The NFL reviews game footage of all major injuries, analyzing each injury frame-by-frame from every angle, recording 150 different variables. The winners’ models automate that process, making review more comprehensive, accurate and 83 times faster than a person conducting the analysis manually.

Insights from the data will be used to inform the NFL’s injury reduction efforts, which include driving innovation in protective equipment design, safety-based rules changes and improvements to coaching and training strategies.

Okay. So the NFL didn’t challenge developers to create AI-powered sensors to detect impacts or machine-learning solutions to help guide medical professionals during on-field incident evaluations.

And it didn’t use algorithms to parse injury scans in order to surface insights into the specific nature of football-related traumas.

The <3 of EU tech

The latest rumblings from the EU tech scene, a story from our wise ol' founder Boris, and some questionable AI art. It's free, every week, in your inbox. Sign up now!

No, what the NFL did was ask people to create algorithms that watch game tape.

Background: In what world is automating safety reviews a good idea? It’s not even useful. Who cares if the AI is 83 times faster at reviewing footage than humans?

It’s not like we’re dealing with millions of hours of footage and struggling to find ways to isolate plays for humans to catch.

When Tesla, for example, trains AI to drive its cars, it has to do so in a simulated space because its not feasible to have millions of agents driving millions of real-world miles. It would take too long and produce too much data for people to view.

But the NFL doesn’t have that problem. Every game tape it produces is watched by millions and reviewed by thousands.

Quick take:The NFL’s responded to its concussion crisis with a non-stop PR blitz. It’s doing everything in its power to make it look like its solving the problem. However, despite it’s best marketing efforts, it’s only managed to reduce the concussion rate by about 25% in the past five years.

Furthermore, the league is worth in excess of $100 billion. If it wants to solve its injury problems, it should invest in real solutions instead of holding dumb contests that serve no purpose other than propaganda.

Nearly 15 million people watch the average NFL regular season game. We don’t need help identifying plays where injuries occur. Players need help mitigating the occurrence of preventable injuries.

Get the TNW newsletter

Get the most important tech news in your inbox each week.

Also tagged with