Tech entrepreneur Sam Altman is launching a new venture, Merge Labs, co-founded with Alex Blania, that is aiming to develop a next-generation brain-computer interface using ultrasound and gene therapy instead of invasive electrode implants. The startup has already recruited biomolecular engineer Mikhail Shapiro—known for his work at Caltech on non-invasive neural imaging—as a founding figure, signalling the firm will pursue a radically different approach from Neuralink. While specifics remain under wraps, funding discussions reportedly involve hundreds of millions of dollars (including from OpenAI) and put Merge Labs at a high valuation ahead of its public launch. The shift toward a “read-only” brain interface that avoids drilling into skulls makes this not only a tech play, but a philosophical statement about augmenting humans via safer, scalable hardware.
Sources: Indian Express, WCCF Tech
Key Takeaways
– Merge Labs is positioning itself as a less-invasive alternative to Neuralink by leveraging ultrasound plus gene therapy for brain-computer interfacing.
– Recruit of Mikhail Shapiro signals serious technical ambition: his work at Caltech on non-invasive neuron-modification and ultrasound imaging lays groundwork for Merge’s mission.
– The venture has already entered substantial fundraising territory (hundreds of millions) and high valuation territory (roughly $800-900 million reported), indicating strong investor confidence in the “human-machine augmentation” thesis.
In-Depth
There’s something striking — and a bit unsettling — about the trajectory of brain-computer interface (BCI) technology: once the purview of sci-fi, now squarely in tech-entrepreneurial territory. With the launch (or imminent launch) of Merge Labs, spearheaded by Sam Altman alongside Alex Blania, we’re witnessing a deliberate pivot away from the skull-penetrating implant model popularised by Neuralink toward something that’s marketed as safer, scalable and more palatable for non-medical use.
The core differentiator? The hiring of Mikhail Shapiro, a biomolecular engineer whose Caltech laboratory has focused on noninvasive neural imaging and control using ultrasound and gene therapy. According to reports, Shapiro has argued that rather than implanting electrodes into brain tissue (which involves invasive surgery, risk of neuronal damage, exclusion of large patient pools and tremendous regulatory burden), it may be simpler to modify cells via gene therapy so they become responsive to ultrasound signals and then use sound-waves or magnetic fields to interact with the brain. Reports say Merge Labs is eyeing just that path. This, in lay terms, means you might one day “think” something and have a machine (or AI) respond, without having a chip drilled into your skull.
On the fundraising front things are already moving fast. Financial Times earlier indicated the company may be valued around $850 million and raising on the order of $250 million (with much of the capital coming from OpenAI’s venture arm) — a pretty bold wager on a startup that hasn’t yet publicly shown product. The Verge columns add that “hundreds of millions” are being talked about and that the announcement is coming in the coming weeks. Those kinds of checks reflect investor belief not only in the technology but in the broader vision of “human-machine merge” — a concept that Altman has long referenced (he wrote in 2017 that many guess humans/machines could merge between 2025 and 2075).
From a conservative perspective, several things stand out. First: the risk profile. Non-invasive may sound safer, but gene therapy plus ultrasound to modulate neurons is still very cutting-edge. There are enormous unknowns: off-target effects, long-term safety, regulatory pathway, and of course ethical concerns about “mind-reading” or “mind-writing” capabilities. Second: the business model. If this technology is truly scalable and non-invasive, the commercial market may be vast—beyond medical (e.g., helping paralysis) into consumer territory (think thought-to-AI interface). But regulatory and public-acceptance hurdles could be formidable. Third: competition and optics. Neuralink has shown human implant trial progress (though slow and controversial); a rival emerging with a different method changes the race dynamic. Investors apparently believe the non-invasive path may win the long game.
Also worth noting is Altman’s own rhetoric: he reportedly said at a dinner he would “definitely not sow something to my brain” that would kill neurons — in reference to the implant-heavy approach. That suggests how Merge intends to position itself in public and regulatory discourse: as the more responsible, “consumer-friendly” BCI startup.
In practical terms for stakeholders — whether investors, regulators, healthcare providers or consumers — Merge Labs is a signal that BCIs are moving from medical intervention to augmentation. That raises questions: Should we consider BCI regulation the next frontier of tech oversight? What standards of evidence, safety and consent should apply if thought-to-machine becomes real? For investors, the space may be becoming a high-stakes version of biotech: huge upside, huge risk, and very long timelines.
From a global-competitiveness angle, the United States – with backing from OpenAI and Silicon Valley ecosystem — is clearly doubling down on hard-tech human augmentation. That has national-security and human-capital implications: if someone can reliably link cognition and machine, how does that impact everything from workforce productivity to military applications? For conservatives, questions of liberty, individual consent, privacy and the human essence matter deeply. A thought interface might amplify human capacity, but who controls the interface? How do we guard against coercion, surveillance, unintended consequences or mission-creep?
In short: Merge Labs may not yet have revealed its product, but its ambition and backing mark a new chapter in BCI technology. The fact that it’s aiming for a less invasive route is notable and may appeal widely, yet the uncharted technical, regulatory and ethical terrain remains vast. If it succeeds, the payoff could be transformative — but the path will be anything but simple.