A federal judge in California has let the core of a privacy class action against Otter.ai proceed, in the first substantive ruling on whether decades-old wiretap law reaches an AI assistant that sits silently in a video call.
The order
Judge Eumi K. Lee issued a 23-page order on 13 August in In re Otter.AI Privacy Litigation (5:25-cv-06911-EKL, N.D. Cal.), granting in part and denying in part Otter's motion to dismiss. The court first rejected Otter's argument that the plaintiffs lacked standing at all.
What survived
The federal Electronic Communications Privacy Act count stands. So does the California Invasion of Privacy Act count, because plaintiffs plausibly alleged a violation of Penal Code §631(a) — the court declining to accept that the Notetaker is a party to the call rather than a third party listening to it. Illinois BIPA voiceprint claims under §15(a) and §15(b) survive, alongside unjust enrichment, UCL and declaratory judgment counts.
What did not
Counts under the CFAA, California's CDAFA and the Washington Privacy Act were dismissed with leave to amend, as were intrusion-upon-seclusion and California constitutional privacy claims as to certain plaintiffs. Plaintiffs have 14 days to amend; Otter must respond within 21 days after that.
The underlying allegation
Plaintiffs say the assistant joins Zoom, Microsoft Teams and Google Meet calls as a silent participant, recording and transcribing medical, financial and professional conversations without all-party consent, and that Otter retains the contents and the voiceprints to train its speech recognition and machine-learning models.
Who else this reaches
Nothing in the reasoning is specific to Otter. Every meeting assistant that joins a call in a two-party-consent state now has a district court opinion saying the "our bot is a participant" defence does not dispose of the claim — and BIPA carries per-violation statutory damages.