Apple FaceID Co-Inventor Builds AI Model for the Human Brain

The co-creator of Apple’s FaceID and Vision Pro has spent half a decade building a frontier AI model designed to decode the brain’s electrical signals, potentially transforming how cognitive disorders are diagnosed. His startup, Hemispheric, just secured $52 million in funding after collecting brain data from 100,000 volunteers to train deep learning systems that can analyze the brain without surgery.

Gidi Littwin left Apple in 2020 seeking a new challenge. That shift came when Hemispheric cofounder Hagai Lalazar reached out to him on LinkedIn. Lalazar had been developing AI to study the brain non-invasively and needed a commercially focused partner. By the time he contacted Littwin, he had interviewed roughly 75 candidates.

At Apple, Littwin helped create FaceID and later worked on hand-tracking for the Vision Pro. That project required gathering what he calls “hundreds of thousands of subjects’ worth of data” to train the deep learning models behind the technology. “There were massive data collection operations behind these projects and we knew we had to build something very similar at Hemispheric,” Littwin says, “and we have.”

Because every person’s brain activity is unique, doctors have traditionally relied on subjective questionnaires and behavioral observation to diagnose conditions like depression, Alzheimer’s, and Parkinson’s. To overcome this, Littwin and Lalazar collected what they describe as their “most prized possession”: 250,000 hours of brain data from 100,000 paid volunteers across Asia, Tel Aviv, and Boston. Participants completed activities that resembled games but activated different brain regions.

This data trained a frontier model that infers brain function from electrical activity inside the skull, much like large language models derive meaning by statistically analyzing text. The team tested the generalized model on subgroups including people diagnosed with PTSD, schizophrenia, and depression, and found it made accurate assessments about their brain health. A clinical study is now underway to see if the model can diagnose and even predict Alzheimer’s.

Hemispheric plans to submit its first product, designed to study PTSD, to the FDA for approval early next year. If cleared, the company hopes to make it available to the public in 2027.

To diagnose a cognitive disorder, a patient wears a lightweight EEG headset that measures brain electrical activity for about 15 minutes while interacting with a tablet app. Hemispheric says its AI will help clinicians decode the signals to make diagnoses, choose the most effective treatment by predicting outcomes, and track progress.

“The future that we envision is one where this is akin to a blood test,” Lalazar says. “The device is going to be very, very cheap; it will be able to be sold and distributed throughout mental health clinics, hospitals, and even psychologists’ offices.”

AI-assisted diagnostic tools for conditions like lung cancer are already in clinical use across Europe, speeding up treatment access. Meanwhile, AI giants like OpenAI and Anthropic are expanding into healthcare, intensifying competition for startups in the space.

Hemispheric’s early-stage funding comes from American and Israeli venture capital firms and individual investors, including early Uber backer Howard Morgan. The company will use the capital to advance partnerships with governments, healthcare organizations, and pharmaceutical firms, hire more staff in the US, and pursue regulatory approval. They also plan to collect brain data from millions more people to improve their model.

The founders are also developing their own brain scanners, which they believe can provide more useful data than traditional EEGs. “These devices were never built for machine learning and definitely not deep learning,” Littwin says.