Hank's IP Brew — Issue #9

Howdy, y'all.

This week: a Villanova legend decides the NCAA's $2.8 billion settlement isn't good enough — and goes it alone. And a group of YouTubers say Apple and Amazon secretly harvested their videos to build AI tools that now compete with the content those tools were trained on.

Let's go.

If you watched the 2016 NCAA championship game, you remember the shot. With 4.7 seconds left, Villanova down one to North Carolina, Kris Jenkins caught a pass at the top of the arc and let it fly. It went in as the buzzer sounded. The Wildcats won the national title. The clip has been replayed millions of times since — in highlight reels, in commercials, in March Madness marketing, in merchandise.

Jenkins never saw a dime of it.

Last spring, he filed a lawsuit against the NCAA and six major conferences, including the Big East, alleging that the organization illegally restricted his ability to earn money from his name, image, and likeness while he was a college athlete. He opted out of the $2.8 billion House v. NCAA settlement — the landmark deal that would distribute damages to Division I athletes over ten years — and decided to pursue his own case instead.

The Athletic ran a fresh feature on Jenkins last week, timed to the tenth anniversary of the shot. His quote sums it up: "I'm happy and appreciative, but I'm also still getting exploited. I was exploited by the NCAA and the university."

Why this matters beyond Jenkins

The House settlement is the largest in college sports history, and the vast majority of eligible athletes will accept it. But Jenkins is one of a small number of opt-outs who believe they can do better on their own — and understanding why is a useful education for any athlete, creator, or anyone else whose content has been monetized without their consent.

When you're in a class action, you give up your individual claims in exchange for a guaranteed (if smaller) payout. The settlement math works in the plaintiffs' favor when individual damages are hard to calculate or when the litigation risk is high. But Jenkins's situation is different. His likeness isn't just valuable in the aggregate — it's tied to one of the most famous plays in the history of his sport. The shot is identifiable, commercial, and has been used repeatedly without compensation. That's a more compelling individual case than the average former athlete who may have some vague NIL-related claim from years of competing.

His lawsuit seeks a declaratory judgment voiding the NCAA bylaws that restricted athlete compensation — not just money damages. That's significant. If Jenkins wins, it could accelerate further structural changes to how college sports handles athlete IP rights, beyond what the House settlement accomplishes.

The takeaway for athletes

Before you sign onto any class action or accept a settlement offer, understand what you're giving up. For most people, the settlement is the right call — certainty beats litigation risk. But if your individual claim is unusually strong — a specific commercial use of your likeness, a clearly identifiable moment, documented revenue tied to your work — it's worth at least asking a lawyer whether opting out makes sense. The window to opt out of the House settlement has closed, but future collective disputes will arise. Know your options before you default to the crowd.

The suspicion has been floating around creator circles for a while: the AI video tools rolling out from Big Tech look a little too familiar. The plant on the desk. The camera angle. The editing style. Creators who've spent years building distinctive YouTube channels have wondered whether their content ended up in someone else's training data — without permission, without credit, and without payment.

Now three of them are doing something about it.

H3h3 Productions, golf creator Matt Fisher (MrShortGame Golf), and Golfholics filed class action lawsuits in early April against Apple and Amazon, accusing both companies of scraping YouTube videos at scale to train their AI models. The Apple suit targets Apple Intelligence. The Amazon suit targets Nova Reel, Amazon's text-to-video generator.

The complaints make specific allegations: that the companies didn't just passively encounter this content — they actively bypassed YouTube's technical protections against bulk downloading, using fake IP addresses, descrambling tools, and virtual servers to extract videos in bulk without detection. And that the financial upside from adding these AI features to Apple devices and Amazon services would not have been possible without the creators' work.

The legal theory runs on the DMCA. Bypassing technical protection measures to extract copyrighted content isn't just a copyright claim — it's a DMCA violation, which carries its own set of potential damages. This is the same statute that protects your content from being ripped directly, and the same one that underlies the takedown system most creators are already familiar with.

The Panda-70M wrinkle

Here's where it gets interesting. The plaintiffs aren't alleging that Apple or Amazon scraped YouTube directly with their own bots. Instead, they're pointing to a dataset called Panda-70M — a massive collection of video clips assembled and distributed by third parties — which the suits allege Apple and Amazon incorporated into their training pipelines. This matters for a few reasons.

First, it means the liability chain runs through an intermediary. If Panda-70M was assembled by scraping YouTube without authorization, then every company that used that dataset potentially inherited the infringement problem. Second, it means creators whose content ended up in Panda-70M may have a claim even if they didn't have any direct interaction with Apple or Amazon.

This is now a pattern. H3h3 and its co-plaintiffs have filed similar suits against Meta, Nvidia, ByteDance, and Snap. These aren't one-off grievances — this is an organized effort to establish that YouTube creators have a cognizable legal right to control how their content is used in AI training, and that companies which bypass technical safeguards to obtain that content at scale can't hide behind the intermediary.

What creators should do right now

You don't need to be h3h3 to care about this. A few practical steps:

Check your YouTube settings. YouTube has introduced controls over whether your content can be used for AI training. Find them and use them. They aren't airtight, but they establish that you didn't consent — which matters if you ever need to prove it in litigation.

Document what's yours. Copyright protects your videos from the moment of creation, but registration gives you access to statutory damages and attorneys' fees. If your channel has substantial viewership, consider registering your most-watched content. The registration fee is small relative to the leverage it creates.

Follow this litigation. These class actions are at an early stage, and a lot will depend on whether courts accept the DMCA bypass theory and what they decide about intermediary datasets like Panda-70M. The outcomes will set the rules for how AI companies can use creator content going forward.

See you next time,

Hank

P.S. If someone forwarded this to you and you find it useful, you can subscribe at newsletter.creatoripacademy.com. No spam, no filler — just IP news that actually matters to creators and athletes.

About Hank's IP Brew

Creator IP Academy helps creators understand and protect their intellectual property. Got a question? Reply to this email.