Apple's Core Argument: Public Access Means No Access Control

Apple’s motion, filed in the U.S. District Court for the Northern District of California, seeks to dismiss the creators’ single claim under DMCA Section 1201(a), the anti-circumvention provision . The company’s central premise is straightforward:

“Plaintiffs voluntarily uploaded their videos to YouTube and made them accessible to anyone who could view the platform, so the videos were not behind a technological access barrier for DMCA anti-circumvention purposes.”

In Apple’s framing, the DMCA prohibits bypassing a lock that restricts access to a work — not restrictions on what happens after a work is already publicly viewable . The plaintiffs, according to Apple, are trying to treat YouTube’s limits on scraping or downloading as DMCA access controls, but public viewability means there was no protected gate to force open .

Apple’s written response reportedly puts it this way: “No password. No payment. No lock. No key.” The company argues that because YouTube streams videos to any visitor without authentication, YouTube’s technical anti-scraping measures do not function as an “access control” under the statute .

The 2026 Wave of Similar DMCA Lawsuits

Apple’s argument does not exist in a vacuum. Throughout 2026, a coordinated group of YouTube creators — led by the same legal team representing h3h3Productions — has filed a series of class actions against major tech firms, all alleging DMCA anti-circumvention violations for scraping YouTube content to train AI models . The table below summarizes the key cases:

The Common Defensive Playbook

The available sources reveal a recurring defense theme across these cases: defendants argue that publicly viewable YouTube videos cannot form the basis of a DMCA access-control claim, because the statute’s anti-circumvention provision targets measures that prevent access to a work, not technical limits on downloading or copying after access has been granted .

• Access vs. Copy Controls: Apple and Snap both argue that the DMCA distinguishes between “access controls” (which restrict who can view a work) and “copy controls” (which restrict what happens after viewing). Their core point: YouTube’s anti-scraping measures are copy controls, not access controls .
• Plaintiffs’ Counter-Theory: The creators allege that Apple deliberately bypassed YouTube’s “controlled streaming architecture” and “rolling cipher” technology — technical measures that do, in their view, regulate access to the underlying video files . They claim that merely being able to stream a video does not mean the public has unrestricted access to download the video file .
• A Critical Precedent: In April 2026, a federal judge in New York denied AI music generator Udio’s motion to dismiss a similar DMCA claim from major record labels, holding that YouTube’s platform protections could qualify as access controls under Section 1201 — a decision that cuts against Apple’s and Snap’s arguments . The court noted that the issue requires further factual development .

Why This Matters: The Future of AI Training and the DMCA

This litigation represents a watershed test of whether the DMCA’s anti-circumvention provisions can be used to police how AI companies obtain training data from publicly accessible online platforms . Two possible outcomes define the stakes:

• If courts accept Apple’s theory: DMCA access-control claims would be largely unavailable against AI training practices that involve public web content. Companies could scrape publicly viewable videos without facing anti-circumvention liability, though standard copyright infringement claims would remain possible .
• If courts accept the plaintiffs’ theory: Bypassing any technical restriction — even one that does not limit viewership — could trigger DMCA liability when used to gather training data. That would impose significant legal risk on any AI company that scrapes publicly available content without authorization .

Early signals are mixed. The Udio decision suggests some courts are willing to let DMCA claims proceed against AI companies . But the Cordova v. Huneault case, which involved a YouTuber suing other creators for using stream-ripping tools, allowed a Section 1201 claim to survive dismissal — a win for the view that YouTube’s rolling cipher is an access control . Meanwhile, Snap’s motion to dismiss, filed in June 2026, is set for a hearing in July 2026, which could provide the first major ruling on these arguments .

As of July 2026, no court has ruled on Apple’s or Snap’s motions in the YouTube scraping cases. The outcome will depend on how courts interpret the line between controlling access to a work and controlling how it is used after access is granted — a question that will define the legal landscape for AI training data for years to come.

Apple's Core Argument: Public Access Means No Access Control

Apple’s motion, filed in the U.S. District Court for the Northern District of California, seeks to dismiss the creators’ single claim under DMCA Section 1201(a), the anti-circumvention provision . The company’s central premise is straightforward:

“Plaintiffs voluntarily uploaded their videos to YouTube and made them accessible to anyone who could view the platform, so the videos were not behind a technological access barrier for DMCA anti-circumvention purposes.”

In Apple’s framing, the DMCA prohibits bypassing a lock that restricts access to a work — not restrictions on what happens after a work is already publicly viewable . The plaintiffs, according to Apple, are trying to treat YouTube’s limits on scraping or downloading as DMCA access controls, but public viewability means there was no protected gate to force open .

Apple’s written response reportedly puts it this way: “No password. No payment. No lock. No key.” The company argues that because YouTube streams videos to any visitor without authentication, YouTube’s technical anti-scraping measures do not function as an “access control” under the statute .

The 2026 Wave of Similar DMCA Lawsuits

Apple’s argument does not exist in a vacuum. Throughout 2026, a coordinated group of YouTube creators — led by the same legal team representing h3h3Productions — has filed a series of class actions against major tech firms, all alleging DMCA anti-circumvention violations for scraping YouTube content to train AI models . The table below summarizes the key cases:

The Common Defensive Playbook

The available sources reveal a recurring defense theme across these cases: defendants argue that publicly viewable YouTube videos cannot form the basis of a DMCA access-control claim, because the statute’s anti-circumvention provision targets measures that prevent access to a work, not technical limits on downloading or copying after access has been granted .

• Access vs. Copy Controls: Apple and Snap both argue that the DMCA distinguishes between “access controls” (which restrict who can view a work) and “copy controls” (which restrict what happens after viewing). Their core point: YouTube’s anti-scraping measures are copy controls, not access controls .
• Plaintiffs’ Counter-Theory: The creators allege that Apple deliberately bypassed YouTube’s “controlled streaming architecture” and “rolling cipher” technology — technical measures that do, in their view, regulate access to the underlying video files . They claim that merely being able to stream a video does not mean the public has unrestricted access to download the video file .
• A Critical Precedent: In April 2026, a federal judge in New York denied AI music generator Udio’s motion to dismiss a similar DMCA claim from major record labels, holding that YouTube’s platform protections could qualify as access controls under Section 1201 — a decision that cuts against Apple’s and Snap’s arguments . The court noted that the issue requires further factual development .

Why This Matters: The Future of AI Training and the DMCA

This litigation represents a watershed test of whether the DMCA’s anti-circumvention provisions can be used to police how AI companies obtain training data from publicly accessible online platforms . Two possible outcomes define the stakes:

• If courts accept Apple’s theory: DMCA access-control claims would be largely unavailable against AI training practices that involve public web content. Companies could scrape publicly viewable videos without facing anti-circumvention liability, though standard copyright infringement claims would remain possible .
• If courts accept the plaintiffs’ theory: Bypassing any technical restriction — even one that does not limit viewership — could trigger DMCA liability when used to gather training data. That would impose significant legal risk on any AI company that scrapes publicly available content without authorization .

Early signals are mixed. The Udio decision suggests some courts are willing to let DMCA claims proceed against AI companies . But the Cordova v. Huneault case, which involved a YouTuber suing other creators for using stream-ripping tools, allowed a Section 1201 claim to survive dismissal — a win for the view that YouTube’s rolling cipher is an access control . Meanwhile, Snap’s motion to dismiss, filed in June 2026, is set for a hearing in July 2026, which could provide the first major ruling on these arguments .

As of July 2026, no court has ruled on Apple’s or Snap’s motions in the YouTube scraping cases. The outcome will depend on how courts interpret the line between controlling access to a work and controlling how it is used after access is granted — a question that will define the legal landscape for AI training data for years to come.