Nine major technology companies, among them Apple, Amazon, Meta Platforms, Microsoft, Nvidia and Samsung, are defending class action lawsuits in Chicago federal court over allegations that they harvested voice recordings without permission to train artificial intelligence models. The complaints were filed by a group of journalists, podcasters and audiobook narrators represented by attorney Ross Kimbarovsky of the law firm Loevy and Loevy, according to a report published on 21 August 2026.
The named plaintiffs include Chicago journalist Carol Marin, Pulitzer Prize winners Yohance Lacour and Alison Flowers, and award-winning journalist Robin Amer. The roster of defence counsel also identifies Alphabet, Adobe and voice AI company ElevenLabs among the defendants, bringing the total to nine companies.
What the lawsuits allege
The plaintiffs contend that the companies collected and used biometric voice information without consent or disclosure, and in doing so violated privacy, publicity, consumer fraud and deceptive practice laws. "They harvested as much information as they could from every source imaginable," Kimbarovsky said.
The complaints do not centre on copyright infringement, although copyright has become a central issue in other legal battles over AI training data. The claims here rest on biometric privacy and consumer protection theories instead.
Why Illinois law matters
The litigation could become an important test of the Illinois Biometric Information Privacy Act, known as BIPA. The statute requires companies to inform workers and consumers before collecting biometric data and to obtain their consent. It also obliges companies to explain how long data will be retained, to maintain a written policy setting out when and how the data will be destroyed, and it prohibits companies from selling user data.
Illinois reduced BIPA penalties in 2024 after a surge in lawsuits invoking the law. Even so, companies remain liable for $1,000 for each negligent violation and $5,000 for each intentional or reckless violation. In a class action covering large numbers of recordings, per-violation damages of that kind can accumulate quickly, which is part of why the technology industry watches BIPA cases closely.
How the companies are responding
The nine lawsuits are currently assigned to seven different federal judges in Chicago. Apple has asked that the cases be consolidated before a single judge.
The defendants argue that the plaintiffs have not shown that the alleged use of their recordings actually occurred, which would mean BIPA may not apply. Apple's lawyers stated: "They allege only that their recordings are available on the internet; they do not offer any facts to support an inference that Apple used those recordings to train its technology."
Meta said it provides people with information about how their data is used and called the plaintiffs' claim incorrect.
The defence teams are drawn from ten major law firms, including Gibson Dunn for Apple, Greenberg Traurig for Samsung, Jenner and Block for Microsoft, Latham and Watkins for ElevenLabs and Meta Platforms, Morgan Lewis for Amazon, Quinn Emanuel for Alphabet and Sidley Austin for Adobe.
What is established and what is merely claimed
Established: the lawsuits have been filed in Chicago federal court; nine cases sit before seven federal judges; Apple has applied to consolidate them; the named plaintiffs and the legal representation on both sides are matters of public record. BIPA's consent, retention, destruction and anti-sale requirements, and its penalty levels of $1,000 and $5,000 per violation following the 2024 reduction, are set out in statute.
Merely claimed: that any defendant actually collected or used the plaintiffs' recordings to train AI models. The defendants say the complaints allege only that the recordings are publicly available on the internet and offer no facts supporting an inference of use, and no court has found otherwise as of 22 August 2026. Whether BIPA applies to voice AI training at all remains untested in this litigation. Meta's assertion that the claims are incorrect is likewise a position statement, not a finding.
Sources
- Apple, Meta, and 7 tech giants face lawsuit over AI voice training data · American Bazaar