Ray Kurzweil, the futurist who has spent decades predicting a merger of humans and machines, is now advising a startup that wants to get there through your nose.
Subsense, a four-year-old neurotech company based in Palo Alto, announced Wednesday evening that Kurzweil has joined as a product and vision advisor. The announcement came at a company event at Subsense's Palo Alto headquarters. According to Fortune, Kurzweil told the room he wants to "merge the smartphone with the brain." That's not new. It's the same singularity thesis he's been pushing since long before Google hired him as a director of engineering in 2012.
What's new is the delivery mechanism. Subsense isn't drilling into anyone's skull. Its brain-computer interface relies on plasmonic and magnetoelectric nanoparticles administered through the nose. They cross the blood-brain barrier and settle into targeted brain regions. From there, a wearable headset with magnetic coils reads neural signals and can stimulate brain activity, all without a single incision. No drilling, no chip, no wires.
No scalpel required #
Compare that to Neuralink, which requires a craniotomy to implant its chip. Synchron threads electrodes through blood vessels instead, less invasive, but still surgical. Subsense founder and CEO Tetiana Aleksandrova has said she wanted something else entirely: the signal accuracy of an invasive system like Neuralink or Blackrock, without cutting into anyone's brain. Her reasoning is blunt: healthy people don't sign up for elective brain surgery. This is her second attempt at building a BCI company. Her first, Neiry, launched roughly seven years ago.
Money has followed the pitch. Subsense has raised $27 million in total seed funding, including a $10 million extension announced in December 2025. The company says the money is going into nanoparticle sensing, in vivo biosafety work, new nanoparticle design and smaller hardware. It's opened a lab and engineering facility at 2445 Faber Place in Palo Alto and is running research collaborations with UC Santa Cruz and ETH Zurich. Pilot human clinical trials are projected to start in 2027.
Fei-Fei Li's World Labs Launches Atlas to Build Walkable AI Worlds World Labs, the spatial AI startup co-founded by Fei-Fei Li, launched Atlas on September 1, 2026, an AI model that builds persistent, explorable 3D worlds from a single photo. The release skips a research paper, pricing, and named partners even as World Labs races Google DeepMind's Genie 3 and Nvidia's Cosmos for the emerging spatial AI category. - how to build 3D worlds from single photographs - AI system that generates explorable walkable 3D environments
The hype question #
Here's the thing worth asking. "Product and vision advisor" is a vague title, and Kurzweil isn't an engineer on this project. He's 78. Fortune described him as a Google researcher, but Subsense's immediate gain is obvious even without a detailed job description: instant name recognition in a field that most people still associate with cyberpunk fiction. When the most famous living futurist shows up at your launch event and repeats his signature line about merging phones with brains, that's a marketing coup, not proof of a technical leap.
That doesn't make the science fake. Aleksandrova's core claim, that nanoparticles can act as tiny antennas moving signals into and out of the brain with something close to invasive-grade fidelity, is a real research bet being tested by real labs, not just a slide deck. Subsense's own technology page says its recent work includes mouse injection results, primary neuron activity imaging with plasmonic nanoparticles and preliminary calcium fluorescence imaging with magnetoelectric nanoparticles. But there's a wide gap between a nanoparticle proving it can carry a signal in preclinical work and a headset reliably reading human thought outside a lab. Subsense hasn't run its pilot human trials yet. Kurzweil's endorsement lands well before there's human data to back it.
Frankly, this is where his track record cuts both ways. Kurzweil predicted the singularity would arrive by 2045 when he wrote The Singularity Is Near in 2005, and he's repeated that timeline for two decades while critics called him a showman more than a scientist. He also has enough standing in AI and computing that Google brought him in to work on machine learning and language processing. Betting on Subsense fits a pattern: he goes where he thinks the biggest leap in human-machine merger is happening next.
For Neuralink, this changes the competitive picture without changing the timeline. Elon Musk's company said in January 2026 that it had 21 Neuralink participants enrolled in trials worldwide, after its first human implant in January 2024. Subsense doesn't have that history. What it has instead is a pitch that skips the operating room entirely, and now an advisor with fifty years of futurist credibility attached to it. Whether nanoparticles delivered through a nasal spray can actually match what a wire threaded into gray matter can do is still an open question, and one that won't get answered until those 2027 trials produce results. Also read: Wall Street Journal Finds Big Tech Hiding $3 Trillion in AI Spending • OpenAI Says GPT-6 Astra Marks the Start of the AGI Era • Conphere Puts AI Agents Behind the Curtain of Event Planning
Founder discussion #
Open in the community → Almost there. Sign in and your reply posts straight away.