Voice creators file BIPA suits against Apple, Amazon, Meta, Microsoft and others over AI voice-training data

What happened

A cluster of lawsuits filed in Chicago federal court pits journalists, podcasters, and audiobook narrators against nine major tech companies over alleged harvesting of voice data to train AI models. Named plaintiffs include Chicago journalist Carol Marin and Pulitzer Prize winners Yohance Lacour and Alison Flowers; they are represented by Loevy & Loevy. Named defendants include Apple, Amazon, Meta Platforms, Microsoft, Nvidia, and Samsung, with ElevenLabs, Google (Alphabet), and Adobe also appearing as defendants with counsel listed. The suits are framed under Illinois' Biometric Information Privacy Act (BIPA), plus publicity, consumer fraud and deceptive-practice claims — arguing the companies collected and used biometric voice information without the consent and disclosure BIPA requires ($1,000 per negligent violation, $5,000 per intentional/reckless violation). The nine cases are currently assigned to seven different Chicago federal judges; Apple has asked for consolidation. Defendants argue plaintiffs have not shown the recordings were actually used to train their technology — Apple's lawyers: "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 informs people how data is used and called the claim incorrect.

Why it matters to creators

Voice is a creator asset: podcasters, narrators, and journalists sell their voices and likenesses. BIPA offers a statutory damages framework independent of copyright fair-use debates, and a test of whether training-data collection counts as biometric collection could give voice creators a new lever. The breadth of defendants — from device makers to model builders — signals how widely voice training data is alleged to be in use.

What is not yet known

These are allegations at an early stage; consolidation is pending and no court has found any defendant actually used the plaintiffs' recordings. Samsung is named as a defendant but is not in the ingest company table and is not primarily an AI model company, so it is noted here rather than attributed.