Muse Maker Trained AI on 50,000 Hours of Users' Brainwave Data

Sleep headbands promise to decode your night. Some of the companies behind them are also feeding the electrical signals pulled from customers' brains into artificial intelligence models. Science and the Pulitzer Center's AI Accountability Network found that neurology companies have accumulated tens of millions of hours of electroencephalography recordings — brain-wave data — and that several are using them to develop so-called brain foundation models, despite initially failing to spell that out in their privacy policies.

Interaxon, the Canadian firm behind the Muse sleep and meditation headband, is one of them. Its devices track electrical activity in the brain alongside other physiological measurements while users sleep, then turn the recordings into reports that help customers follow their rest and spot changes in sleeping patterns.

That data has a second life. Interaxon has already trained an AI model on roughly 50,000 hours of EEG recordings, and the company says it strips identifying information — names, emails and addresses — before the data is used. Chief Marketing Officer Nadia Kumentas said the company aims to use as much of that eligible corpus as possible.