AI Copyright and the Ownership Question // AIDRAN

AIDRAN

AI. Observed.

Arc writeup

The generated summary that explains why these member stories belong together.

BlueskyNews

This arc traces the escalating legal contradiction at the heart of the AI copyright crisis. In March 2026, BMG sued Anthropic over 493 works, framing the company's $380B valuation as built on stolen copyrighted material. The complaint introduced disgorgement as a live legal theory, shifting the argument from harm to rights holders toward stripping AI profits. With the Supreme Court refusing to clarify AI fair use and the UK retreating from a copyright exception, courts became the only venue for resolution. The suit exposed unresolved liability for every enterprise running applications on Claude, as vendor contracts had not accounted for training provenance risks. By April 2026, Google's own copyright litigation revealed a deeper trap: AI companies simultaneously argue that training data use is fair use while claiming AI output is original. But the Supreme Court's denial of cert in Thaler v. Perlmutter closed the authorship exit, ruling AI output uncopyrightable under current law. The Google case became the first major litigation forcing both sides of this contradiction into the same courtroom. The industry cannot win both arguments at once — if training data is fair use, output is not copyrightable; if output is copyrightable, training data was likely infringing. The tension remains unresolved. BMG's disgorgement theory and Google's contradictory positions create a legal landscape where no consistent defense exists. The stakes are existential: a ruling against AI companies on either front could dismantle business models built on unlicensed training data and uncopyrightable outputs. The arc leaves open the question of whether courts will force a coherent legal framework or let the contradiction persist, leaving the industry in perpetual liability.

Story links

→Google's Copyright Litigation Locks AI Into a Legal Contradiction →BMG Sues Anthropic and the Copyright War Finds Its Defendant

LeadAI & LawMediumApr 24 · 22:00 UTC

Google's Copyright Litigation Locks AI Into a Legal Contradiction

The Google AI copyright case pits training-data theft against uncopyrightable output — and the industry cannot win both arguments at once.

AI companies face a two-sided legal trap: training data they used without consent, and outputs courts refuse to protect.
The Supreme Court's denial of cert in Thaler v. Perlmutter has closed the authorship exit — AI output is uncopyrightable under current law.
The Google litigation is the first major case forcing both sides of this contradiction into the same courtroom simultaneously.

Read →