Data vs. the People
Meme
The Supreme Court in a Quiet Decision Just Ruled on IP Ownership Who Owns the Internet if Everyone Stole it at Once?
On April 17, 2025, the U.S. Supreme Court issued a 5–4 ruling in Data v. Everyone—a case now widely referred to as Data vs. the People.
At stake was a single question that would have sounded impossible a decade ago:
Can first-mover theft serve as a legal defense in the age of AI?
The Court’s answer: yes.
When multiple AI developers have used the same unlicensed content, ownership belongs to whoever stole it first, built with it fastest, and got caught last.
These factors need to be weighed "carefully and thoughtfully," the court said in its majority opinion.
This newly established doctrine—Prior Unauthorized Use (PUU)—formalizes a quiet reality: In generative AI, scale beats consent. Speed beats attribution.
The Industry’s Response: Calm, Confident, and Unapologetic
Following the ruling, OpenAI released a statement:
“We stole the content in good faith, with no prior knowledge that anyone had stolen it first. We have a clean conscience.”
Google was more enthusiastic:
“We have more experience in this realm than anyone. What amazes us is that it took this long to normalize this.”
Meta claimed historical justification, citing ingestion systems predating regulatory frameworks and that it had "tried to cover it up for as long as possible and it shouldn't be blamed for getting caught."
The Legal Innovation: Pre-Stolen Ambiguity
Many of the datasets in question—including Reddit threads, Wikipedia entries, YouTube transcripts, and digitized books have passed through so many paraphrasers, scrapers, APIs, and aggregation layers that their origins are no longer traceable.
This state is now referred to as Pre-Stolen Ambiguity:
The Copyright Office has declined to take action, stating that unless a human author can demonstrate a “molecular correlation between authored content and model output,” no claim can proceed.
Meanwhile, AI companies are publishing self-certified ingestion ledgers, backdated provenance statements, and calling it transparency.
Legislative Proposals Are on the Table
A memo from the House Subcommittee on Algorithmic Appropriation has proposed:
A federal Data Lineage Registry
Blockchain timestamping for future ingestion
Voluntary disclosure protocols for model builders
Redefining ownership based on integration timing, not originality
Still, the memo makes one thing clear:
Nothing proposed will apply retroactively.
Stephen Klein is Founder and CEO of Curiouser.AI, the only Generative AI designed to augment human intelligence, not replace it. He also teaches AI Ethics at UC Berkeley. To learn more visit curiouser.ai
