Filed June 30, 2026 in the Western District of Washington at Seattle (No. 2:26-cv-02281) by Eric Anderson, derivatively on behalf of Microsoft Corporation, through Cotchett, Pitre & McCarthy. See Anderson v. Microsoft (copyright shareholder-derivative suit).
Defendants
Named individually: Satya Nadella, Amy E. Hood, Jared Spataro, Rajesh Jha, Reid Hoffman, Hugh Johnston, Teri List, Catherine MacGregor, Mark Mason, Sandra E. Peterson, Penny Pritzker, Charles W. Scharf, John W. Stanton, and Emma Walmsley. Microsoft Corporation is the nominal defendant — the structure of a derivative action, in which the corporation is the party in whose right the claim is brought.
The claim
Two counts: breach of fiduciary duty and violation of Section 14(a) of the Securities Exchange Act of 1934, with a jury demand. The relevant period is stated as January 1, 2022 to the present.
The alleged wrongdoing has two halves. Defendants "caused Microsoft to make false and misleading statements about its artificial intelligence strategy, the Company's Copilot family of products, and financial results." And during the same period they "caused the Company to violate copyright and intellectual property laws by training its AI software on copyrighted works for which it did not possess lawful licenses."
The disclosure theory
What distinguishes this from the direct copyright suits is that the alleged wrong is the statement, not the training. The complaint quotes Microsoft's representations to shareholders and the market:
"Generative AI Models — Microsoft uses a variety of data sources, including publicly available information, in a manner consistent with global copyright laws."
"we do not train on data from domains listed in the Office of the United States Trade Representative Notorious Markets for Counterfeiting and Piracy list"
"We train on data that we gain access to (such as archives and metadata) through negotiated arrangements with publishers and copyright owners."
Against these it sets the litigation record: "Beginning in September 2023, however, Microsoft was sued for copyright infringement by numerous authors and copyright holders. The first of such cases was Authors Guild v. OpenAI & Microsoft Corp., No. 1:23-cv-08292 (S.D.N.Y.)" — described as "the first major class action against Microsoft" of this kind.
The theory therefore does not require proving infringement. It requires showing that the assurances were misleading when made and that directors were on notice — which is why the September 2023 filing date is load-bearing against a relevant period beginning January 2022.
Standing of these allegations
A complaint states allegations, not findings; Microsoft's response is not in this document, and the copyright suits it relies on were themselves unresolved at filing. What it establishes is the emergence of a second-order liability theory: that training-data provenance creates securities and fiduciary exposure through what companies tell investors about it, independent of how the underlying copyright questions resolve. See AI Copyright.
Relationships
- supports: Anderson v. Microsoft (copyright shareholder-derivative suit) — the primary text behind that page
- related: NYT v. Microsoft, OpenAI et al. — one of the underlying infringement actions the derivative theory rests on
- related: AI Copyright, AI and Tort Liability, Microsoft, OpenAI