Killing the Internet: Why Live Sports Content Will Reign Supreme
Generative AI is showing the potential to transform productivity and efficiency…
But it may also kill human-made creativity and incentives on the internet.
Interestingly — I think this dynamic is good for sports.
Let’s Dive In 👇
From Private to Public
Reddit recently went public at a valuation of $6.5B (its shares are currently up % at $47).
The social media giant’s IPO reemphasized that tech companies care more about their profits than the people creating content and making the platforms worth anything.
These points help illustrate the insanity:
Reddit CEO Steve Huffman made $193 million in 2023 while laying off 90% of the staff
None of the moderators or content creators made a penny from Reddit’s public listing
Google paid Reddit $60 million to train its AI models on Reddit's posts
When asked about taking the company public, Huffman mentioned “ownership, community, and data that grows with more users” — all things they took advantage of and sold to the highest bidder.
PR persona Ed Zitron believes that:
We're at the end of a vast, multi-faceted con of internet users, where ultra-rich technologists trick their customers into building their companies for free.
Tech executives see users as “veins of data” to be exploitatively mined as many times as possible, given nothing in return other than access to a platform that works properly.
And all of this human-generated content across the internet is now being copied at a mass scale.
Feeding the Machine
Ultimately, executives think they've found a way to replace human beings making cool things with generative platforms trained on datasets controlled and monetized by trillion-dollar firms.
So how does this work?
Generative AI models are trained by using massive amounts of text scraped from the internet.
But there’s a problem on the horizon — as more internet content is created, either partially or entirely through generative AI, the models themselves will start training themselves on content written by their own models.
So what does this mean?
The consumer adoption of generative AI has brought a degree of radioactivity to its own dataset — machines training machines on their own content.
Hacking the Algorithm
Social platforms were built to reward scale and volume far more often than quality.
AI naturally rewards those who can find the best ways to manipulate the algorithm.
What we’re seeing:
Spammers making thousands of dollars from TikTok's creator program by making "faceless reels" where AI-generated voices talk over spliced-together videos ripped from YouTube.
AI-generated content dominating bestseller lists, forcing Amazon to limit authors to publishing three books a day.
Etsy pumped full of AI-generated art, t-shirts, and mugs that use ChatGPT to optimize listings to rank highly in Google search.
Job applications, magazines, and media sites flooded with AI-generated essays.
The eventual goal of these massive platforms (which control the data) seems to be to standardize the internet and turn it into a series of toll roads that all lead to the same place (so that they can control and monetize the results as much as possible).
So why is it working?
Generative AI is a perfect tool for soullessly churning out content to match a particular set of instructions — such as those that an algorithm follows.
"Viral" content is no longer a result of lots of people deciding that they find something interesting — it's a condition created by algorithms manipulated by Gen AI forces.
A huge problem is that it’s incredibly difficult to make a model forget information and with dozens of copyright lawsuits in the open — who knows what could happen.
My Viewpoint
While web3 is now in the backseat — the essence of a blockchain makes perfect sense.
“Data owned by the users with privacy and transparency in place”.
I think it’s feasible to suggest that many of the blockchain companies knew what was coming with generative AI.
The pendulum is sure to swing back — and I think we’ll see a reshuffling with more niche communities/pockets of the internet.
Many people (myself included) no longer trust building a brand or the safety of their data on any of these tech platforms.
The platforms are getting diluted with AI content (e.g. about 25% of the comments on my LinkedIn posts are clearly someone using ChatGPT).
Humans want human content — not machine-led algorithm-hunting content.
Where’s This Leave Us
Gen AI will transform music, TV shows, movies, and other forms of creative art.
But sports are…
One of the only industries with built-in communities.
The only programs watched in real time anymore (and advertisers are hungry for more).
You can’t fake sports content.
Hence why TV deals for emerging leagues are likely to be much larger than the current ones (and worth adding as upside arbitrage).
Generative AI has its place and will certainly transform internal operations within sports (as they are very data-heavy).
But it will be impossible to recreate actual sports themselves.
Sports may truly be the last form of human-content ever to exist.
Podcasts 🎙
Awesome guests this past week:
Ari Kaplan - Head of Evangelism, DataBricks (listen here)
Often dubbed “The Real Moneyball Guy”, we dive deep into the intersection of sports, artificial intelligence, and data analytics.
Vaidhy Murti - CEO/Founder, WIT Sports (listen here)
We explore his lifelong entrepreneurial journey, why he’s building in sports, and how brands are thinking about digital activations.
I appreciate you reading today!
Something Ari said in the podcast still has me stumped “AI becomes real when it passes the Turing Test, but we won’t even be able to tell by that point”.
Peace,
AP
Sources:
https://www.cnbc.com/2024/03/20/reddit-prices-ipo-at-34-per-share-sources-say.html
https://www.bizjournals.com/sanfrancisco/inno/stories/news/2024/02/23/reddit-ceo-steve-huffman-total-compensation-2023.html
https://www.wheresyoured.at/are-we-watching-the-internet-die/
https://www.404media.co/inside-the-world-of-tiktok-spammers-and-the-ai-tools-that-enable-them/
https://arstechnica.com/information-technology/2023/09/ai-generated-books-force-amazon-to-cap-ebook-publications-to-3-per-day/
https://www.theatlantic.com/technology/archive/2023/06/ai-chatgpt-side-hustle/674415/
https://www.wsj.com/articles/chatgpt-already-floods-some-corners-of-the-internet-with-spam-its-just-the-beginning-9c86ea25?