Recommended Articles
Fundraising

Sabi Raises $50M Seed to Read Brain Signals Through a Baseball Cap

By Angeled · October 11, 2026 · Fundraising
Sabi Raises $50M Seed to Read Brain Signals Through a Baseball Cap
Sabi Raises $50M Seed to Read Brain Signals Through a Baseball Cap

On October 9, 2026, Sabi announced a $50 million seed round led by Khosla Ventures, with participation from Accel, Initialized Capital, former OpenAI chief product officer Kevin Weil, DST Global, Collaborative Fund, and Ascend. The raise is a substantial seed bet on a consumer-grade brain-computer interface that requires no surgery, no gel, and no scalp contact.

The Product: Neural Sensing in Disguise

Sabi's first product looks like an ordinary baseball cap. Hidden inside are sensors designed to read neural signals through hair — across a gap of up to 5 millimeters — using sensors just 1 to 5 millimeters in size. The system is built around a custom EEG chip, and the company says it has accumulated more than 100,000 hours of labeled neural recordings, which it describes as the largest disclosed dataset of its kind.

On an internal benchmark, Sabi claims its chip is 50 times more power-efficient than the Texas Instruments ADS1299, a widely used analog front-end chip for biopotential measurement. A prototype is expected to debut at CES 2027.

The stated goal: translate imagined speech and user intent into commands for AI systems — a private, potentially faster input method than typing or speaking.

The Technology: EEG, Upgraded

Electroencephalography (EEG) — reading electrical activity from the scalp — is one of the oldest non-invasive brain-sensing methods, in clinical and research use for nearly a century. Its historical limitations have been well known: signals are faint, spatial resolution is coarse, and traditional setups require conductive gel and precise electrode placement.

Sabi's claimed advances attack each of those problems: dry sensors that work through hair, a custom chip for power efficiency, and a large labeled dataset for machine-learning decoding. The through-hair sensing is the key consumer unlock — nobody will wear a gel-and-wires headset casually, but a baseball cap is already socially invisible.

Contrast With Clinical BCIs

The brain-computer interface field is usually discussed in terms of implantable systems — surgical devices that place electrodes directly on or in the brain for precise, high-bandwidth readings, typically developed for clinical use in patients with paralysis or severe motor impairment. Those systems offer the highest signal quality but face regulatory, surgical, and adoption barriers that keep them out of consumer markets.

Sabi is placing its bet on the opposite end of the spectrum: lower-fidelity but vastly more accessible sensing, paired with AI models that do more interpretive work. If the decoding works well enough for intent and imagined speech in everyday settings, the addressable market is everyone with a head — orders of magnitude larger than the clinical population.

Why It Matters for Founders and Investors

For founders, Sabi illustrates the "consumerize the lab instrument" playbook: take a decades-old sensing technology, shrink it, remove the friction (gel, wires, scalp contact), and pair it with modern machine learning. The $100,000-hour dataset claim is the moat thesis — hardware can be copied, but large labeled neural datasets are hard to replicate. Founders in health-tech and wearables should note how much of this pitch is really about data, not just sensors.

For investors, the round is a bet that AI input methods are the next interface war. Voice, typing, and touch all have limits; neural input could be faster and private. But the technical risk is real — internal benchmarks and claimed dataset sizes aren't independently verified, and the gap between prototype and reliable consumer product in neurotech is historically wide. Khosla's lead signals confidence; the CES 2027 debut will be the first public test.

Angeled.com connects founders with investors.

Read Next