Alphabet shareholder sues Pichai and board over alleged unlawful AI training
The derivative suit seeks damages for Alphabet—not the shareholder—and alleges that executives exposed the company to copyright and voice-privacy liability.
Loading page…
The derivative suit seeks damages for Alphabet—not the shareholder—and alleges that executives exposed the company to copyright and voice-privacy liability.
Listen to this story
Filed in Chicago federal court on October 30, 2026, Phil Rosen’s derivative suit asks defendants to compensate Alphabet, repay executive compensation and strengthen copyright and biometric-privacy oversight; Rosen seeks no personal payout and specifies no damages amount. The complaint alleges Google trained music and voice AI on copyrighted recordings and speech collected without consent, while Alphabet misrepresented its controls in proxy filings. It links those alleged disclosure failures to $169.6 billion in share repurchases—not to a damages demand. The claims have not been established in court.
For MusicLM, Rosen alleges Google used five million audio clips totaling 280,000 hours without identifying recordings, rights holders or licenses.
The complaint names Lyria, Lyria 2, Lyria 3, Mulan and MusicLM among systems allegedly developed using copyrighted music.
Rosen alleges Gemini Live used hundreds of thousands of hours of speech without the written notice and release required under Illinois’s biometric privacy law.
Alphabet’s board told investors it had oversight and data-tracking systems for responsible AI development. Now shareholder Phil Rosen is suing Sundar Pichai and 12 other directors and executives, alleging they knowingly allowed unlawful training practices while misleading investors. Filed October 30, 2026, in Chicago federal court, the case seeks compensation for Alphabet itself.
Rosen’s 79-page complaint is a shareholder derivative suit: he is suing on the company’s behalf, rather than seeking a personal payout. Music Business Worldwide obtained the filing. Alongside Pichai, the defendants include cofounders Larry Page and Sergey Brin, Chair John Hennessy, CFO Anat Ashkenazi and chief legal officer Kent Walker.
The complaint accuses Alphabet’s leaders of breaching their duties to the company by knowingly building AI products on pirated books, artwork and music, and collecting voiceprints without consent. Bloomberg Law reports that Rosen also alleges misleading disclosures in annual filings. These are allegations, not findings established in court.
Rosen does not specify a damages amount. He asks the court for three forms of relief:
The strategy was carefully crafted to avoid spending the time and money necessary to obtain lawful licenses or permission from the copyright holders, and instead to simply pay damages later when Alphabet was sued by the copyright holders and commercial voiceprint holders.
Phil Rosen’s complaint, as quoted by Music Business Worldwide
The music allegations focus on YouTube. Rosen claims Google copied millions of copyrighted sound recordings, compositions and lyrics without permission or payment. The complaint names Lyria, Lyria 2, Lyria 3, Mulan and MusicLM as systems developed using those works.
It describes a process of extracting audio, converting it into machine-readable formats and cutting it into training clips. Rosen also alleges that Google removed information identifying artists and copyright notices. For MusicLM, the complaint cites five million clips totaling 280,000 hours, saying Google did not identify the recordings, rights holders or licenses in that collection.
Much of this music section draws on independent musicians’ proposed class action against Google, filed in the same court in March. Some passages repeat that case’s wording. Google has sought dismissal of that lawsuit, arguing that the musicians accepted YouTube terms granting a license extending to affiliates, including Google.
The voice allegations invoke a different law. Rosen claims products including Gemini Live were built using hundreds of thousands of hours of speech without consent required by Illinois’s Biometric Information Privacy Act, or BIPA. That law requires written notice and a written release before collecting biometric data such as voiceprints.
Rosen’s disclosure claims point to specific board statements. Alphabet’s 2024 proxy said directors received regular updates from managers implementing its AI principles. The complaint says those assurances disappeared from the 2025 and 2026 proxies, and alleges the earlier language falsely suggested compliance with copyright and biometric-privacy laws.
The 2025 proxy contained another assurance Rosen disputes. Opposing a shareholder proposal for an AI data-usage report, the board described an auditable system tracking AI assets, data origins and usage. Rosen alleges that statement was false and misleading.
He ties those alleged disclosure failures to a financial consequence: $169.6 billion in share repurchases. The complaint claims undisclosed copyright and BIPA violations inflated Alphabet’s stock price, causing the company to overpay when buying back its own shares. That figure is the alleged buyback spending, not a damages demand.
Loading discussion...
Join the conversation
Explain where board responsibility should begin and end.
Be the first to share a perspective or an experience.
Reader comments
Newest comments first. Replies stay oldest first.