In the rapidly evolving world of brain-computer interfaces, a new player has emerged from the shadows of neuroscience research, carrying with it a promise that feels plucked from the pages of science fiction. Merge Labs, which officially spun out of the neurotechnology company Forest Neurotech this past January, is not just another startup dabbling in the field; it is a collective of visionaries determined to make reading the human mind as simple as putting on a pair of headphones. The company was co-founded by a remarkable group of individuals, including Norman and Aflalo, who are joined by none other than Sam Altman, the tech entrepreneur and investor known for his role at OpenAI, along with researcher Mikhail Shapiro, and entrepreneurs Alex Blania and Sandro Herbig. The very presence of such diverse and accomplished founders signals that Merge Labs is aiming for something big, something that could redefine how we interact with machines. At its core, the company’s mission is deceptively straightforward: to create a brain-computer interface (BCI) that is safe, effective, and, crucially, noninvasive. While many BCI companies focus on surgically implanted electrodes that read the brain’s electrical signals directly, Merge Labs is taking a different path, one that uses ultrasound to listen to the brain’s inner workings from the outside. It’s a technology that has been gestating in research labs for years, and now, with the formation of Merge Labs, it is stepping into the spotlight. The founders envision a world where thought alone can control devices, where people with severe paralysis can communicate effortlessly, and where the barrier between human intention and digital action simply melts away. But as with any ambitious endeavor, the journey is fraught with scientific and engineering challenges. To understand what makes Merge Labs so intriguing, and so promising, we must first delve into the science behind their approach, the hurdles they face, and the quiet, methodical way they are going about changing the world.
The technological heartbeat of Merge Labs lies in the use of ultrasound to measure brain activity indirectly, through the flow of blood rather than the firing of neurons themselves. This is a crucial distinction that sets it apart from the electrode-based systems that have dominated the BCI landscape. When neurons in the brain become active, they require more oxygen and nutrients, which are delivered via the bloodstream. This results in a localized increase in blood flow and volume, a phenomenon known as the hemodynamic response. Ultrasound waves, which are high-frequency sound waves, can be aimed at specific regions of the brain, and the echoes that bounce back provide a real-time map of this blood movement. It is a brilliant, elegant workaround, but it comes with an inherent limitation: the delay. As Dr. Maryam Shanechi, a neuroengineer at the University of Southern California, points out, there is a natural lag of a few seconds between the moment neurons fire and the subsequent change in blood flow. “It captures a slower variation in neural activity,” she explains. “That’s fine for a lot of applications, but it depends on what you want to build.” This is not a trivial concern. For tasks that require lightning-fast reactions, such as moving a robotic limb with the fluidity of a natural arm or typing at a rapid pace on a virtual keyboard, this few-second delay could be the difference between a tool that feels magical and one that feels frustratingly sluggish. The brain operates on a timescale of milliseconds; a system that lags behind by whole seconds would break the illusion of intuitive control. Recognizing this, the team at Merge Labs is not simply accepting this limitation as an immutable fact of nature. Instead, they are planning to use predictive AI models to bridge the gap, anticipating what the brain is going to do before the blood flow signal fully materializes. These machine learning algorithms will learn to recognize patterns in the ultrasound data, allowing the system to extrapolate forward and reduce the perceived lag. It is a classic engineering problem, one that is being solved with cutting-edge artificial intelligence, and it speaks to the company’s pragmatic, forward-thinking approach. While Biederman, one of the co-founders, declined to comment on specific plans, he acknowledged that Bridge (as the company’s technology is sometimes called) is interested in both consumer and medical applications. The potential is enormous, from helping locked-in patients communicate to creating entirely new ways of interacting with video games and virtual worlds.
The company is taking its first, cautious steps toward making this vision a reality, having just launched its first clinical study in Redwood City, California. This is an observational study, meaning that it is not testing any treatments or diagnoses, but rather gathering crucial baseline data about how the ultrasound technology performs on human heads. The study is carefully designed to include a diverse group of participants: healthy adults, as well as those who have undergone skull surgery, which often creates natural acoustic windows that are more transparent to sound. In these sessions, a small ultrasound probe is placed on the side of the head, and participants are asked to perform a series of tasks—listening to music, watching videos, moving their arms, or speaking aloud. Meanwhile, the probe, which is built using semiconductor silicon chips made by the portable ultrasound company Butterfly Network, records the changing patterns of blood flow in the brain. This is not the first time ultrasound has been used to read brain activity; indeed, it has been a powerful tool for researchers for decades. But what makes this study different is the form factor and the ambition. The probe used in the study is a far cry from the bulky, room-sized ultrasound machines of the past. It is small, portable, and designed to be worn, bringing us closer to the dream of a truly wearable BCI. The choice of Butterfly Network as a partner is also telling; it signals that Merge Labs is serious about making its technology accessible and affordable, leveraging the advances in semiconductor technology that have made ultrasound as small as a computer chip. The study itself is a testament to the methodical, no-shortcuts approach the company is taking. Before they can even think about selling a consumer product, they need to understand the nuances of the ultrasound signal, how it varies from person to person, and how to filter out the noise of everyday movement. It is painstaking, unglamorous work, but it is the foundation upon which everything else will be built.
Perhaps the most striking aspect of Merge Labs’ technology is its physical form. While the company is not yet releasing images of its device, Biederman describes it as “something close to headphones” that will sit on the temporal window, a region of the skull located slightly above the ear. This is no arbitrary choice. The temporal window is the thinnest part of the skull, making it the easiest point through which to image the brain. “It’s the easiest place to image through,” Biederman says. This is a matter of simple physics: bone scatters and absorbs sound waves, so the less bone the ultrasound has to pass through, the clearer the signal. This is a critical insight, because it highlights one of the biggest technical hurdles in the entire field of noninvasive brain imaging. The skull is a formidable barrier, and many researchers have spent their careers trying to figure out how to peer through it. For Merge Labs, the temporal window is the key that unlocks the whole enterprise. By focusing on this specific location, they can achieve a level of signal quality that would otherwise be impossible without surgical intervention. But even with this advantage, the challenges are immense. A 2024 study from researchers at Caltech and the University of Southern California demonstrated that an ultrasound-based BCI could read and decode human brain activity, but it did so by requiring a portion of the skull to be removed and replaced with a transparent “window” through which the ultrasound waves could pass. While this approach is highly effective, it is also highly invasive, and it is precisely the kind of procedure that Merge Labs wants to help people avoid. The company’s entire philosophy is centered on the idea that brain-computer interfaces should not require brain surgery. They believe that the benefits of a BCI—the ability to communicate, to control devices, to regain lost function—should be accessible without the risks and complications of an operating room. This is a bold stance, and it is not without its skeptics. Some researchers believe that noninvasive systems will never be able to achieve the same level of precision and speed as their invasive counterparts. But Merge Labs is betting that with the right combination of hardware, software, and artificial intelligence, they can come close enough to make a real difference in people’s lives.
Ultimately, the story of Merge Labs is a story about balance—the balance between speed and safety, between ambition and realism, and between the wonders of technology and the practicalities of human biology. The company’s journey is still in its early stages, but the path forward is becoming clearer. Biederman’s vision is focused on noninvasive interfaces because he believes that is where the biggest opportunity to help people lies. He is not alone in this belief. The field of brain-computer interfaces is at a critical inflection point, with companies and researchers around the world racing to develop systems that can decode thoughts with ever-greater accuracy. What sets Merge Labs apart is its willingness to look beyond the electrical signals that have dominated the field and to embrace a different kind of signal, one that is slower but also safer, one that does not require drilling holes in the skull but can instead be delivered through a device that resembles a pair of headphones. The road ahead will be long. There will be setbacks, failures, and difficult lessons learned. But the potential reward is nothing less than the ability to liberate the human mind, to give a voice to those who have been silenced by illness or injury, and to fundamentally change the way we interact with the digital world. As the clinical study in Redwood City gets underway, and as the team at Merge Labs continues to refine their technology, one thing is certain: the future of brain-computer interfaces is being written right now, and it is being written with sound waves, not just electrodes. The world is watching, and the possibilities are nothing short of breathtaking.
