Helix Raises $6 Million for On-Device Voice Verification

Helix has raised $6 million in funding, in a round led by Chicago Ventures, according to TechCrunch, which reported the round on November 11, 2026.

The round also drew participation from BAM Ventures, Aglaé Ventures, Corazon Capital, Tusk Ventures, Yonder Ventures, Rule30 and Brown Alumni Group. Dag Kittlaus, a co-founder of Siri, backs the company and will serve as an advisor. Helix was founded in 2025 by Alex Oberg, Anna Dorofiyenko and Mike Hall. Oberg previously worked at Ticketmaster, Dorofiyenko led data science teams at Live Nation and Zillow, and Hall worked at Nike and Tinder.

The company's core voice technology runs directly on a user's device rather than on a remote server. It determines whether a speaker is human and estimates that speaker's age. Helix said its model can run locally even on an iPhone 13.

Helix says the technology can replace or supplement CAPTCHA challenges, the image and text tests websites use to screen out bots, and help block bot attacks. Zeel, an in-home healthcare provider, already uses it to verify the identities of providers visiting customers' homes. Helix is also running pilots to explore whether voice can serve as a biomarker, a measurable signal of the body's state, for detecting drunkenness or distress.

The voice technology does not identify individual speakers. Helix said it cannot reliably detect whether someone is lying, and that it is not yet capable of reliably detecting AI agents during live calls.

The broader context here is where the verification happens. Running inference on the handset, rather than in a cloud endpoint, changes the latency profile of a liveness check and changes what has to leave the device. A system that answers two narrow questions, whether the speaker is human and roughly how old they are, is a different object from a biometric identity service. The design choice points toward the narrower product, and the company's own disclaimers reinforce that reading. The live-call gap matters for the same reason, since a human-versus-machine check that cannot yet flag an AI agent on a live call sits at the edge of the problem it is most often sold to solve.

In my view, the healthcare and biomarker items are the least settled parts of the announcement. The Zeel deployment is a concrete use: a care agency needs to confirm that the person arriving at a client's door is the credentialed provider it dispatched. Detecting drunkenness or distress from voice is a different claim, one that depends on validation data the announcement does not describe. Pilots are the correct way to test such a claim, and they are also where overreach tends to start.

Taken together, the CAPTCHA use case carries the largest user footprint, and it is also the one most exposed to adversaries improving at imitating humans. If an on-device check can separate a live human voice from bots at low cost, and if it works on hardware already in many people's pockets, a familiar point of friction could shrink for ordinary users. Whether it holds up under sustained attack is a question for the next round of deployments, not for a funding announcement.

For now the facts are these: a $6 million round led by Chicago Ventures, a 2025 founding team with product and data backgrounds, a Siri co-founder in an advisory role, and a voice system that runs locally and is explicitly scoped to human detection and age estimation.