Episode Summary
Executive Summary: Max Hodak argues brain-computer interfaces are entering a takeoff era, with Science already restoring partial vision in blind patients via the PRIMA retinal implant and pursuing biohybrid and perfusion programs. He frames BCI as a broad category like pharma, spanning restoration, enhancement, and eventually ultra-high-bandwidth human-machine integration, while emphasizing plasticity, engineering over drug discovery, and a 2035 horizon of major medical and cognitive change.
Main Topics: PRIMA retinal prosthesis and restoring vision (Priority: 5/5): Science’s lead product is a retinal implant that bypasses dead rods and cones by projecting images into a tiny silicon chip under the retina, enabling blind patients to perceive black-and-white form vision and read chart letters. BCI as a platform, not a single product (Priority: 5/5): Hodak argues BCI will resemble pharma: many modalities, probes, and applications rather than one universal device, with different approaches serving vision, hearing, motor control, sleep, focus, and future enhancement. Neuroplasticity and rehabilitation (Priority: 4/5): The conversation explores how the brain adapts to feedback, why critical periods matter, and why adult brains still show meaningful plasticity, including learning to control implanted signals within minutes. Biohybrid neural interfaces and consciousness (Priority: 4/5): Hodak describes a long-term vision for living-neuron grafts that could form new neural connections, potentially enabling high-bandwidth brain-to-brain links and new forms of consciousness or phenomenal binding. Engineering-first medicine vs. drug discovery (Priority: 4/5): He contrasts neural engineering with traditional drug discovery, arguing that for many neurological and sensory problems, engineered devices can outperform slow, uncertain molecular approaches. Science’s broader pipeline: retina, neural interfaces, perfusion (Priority: 3/5): Beyond vision, Science is building neural interface and vessel/perfusion technologies, motivated by the gap between what ICU/organ-support tech can do and what is economically and clinically deployable. Startup lessons and the origin story of Neuralink/Science (Priority: 3/5): Hodak reflects on his software background, high-agency path into BCI, lessons from Neuralink’s execution, and how first-principles thinking plus elite teams accelerated progress.
Key Arguments: BCI is not one product but a category of technologies, like pharma, with different modalities for different use cases. Restoring lost function is the most immediate and ethically compelling BCI market because risk-reward is strongest in severely disabled patients. The brain remains highly plastic under feedback, so devices can be learned and adapted to, even in adulthood. Science’s retinal prosthesis is notable because it creates coherent form vision rather than just phosphenes or flashes. Engineering the nervous system may be more effective than drug discovery for many conditions, especially where biology is hard to target precisely. Biohybrid interfaces may eventually create much higher-bandwidth communication than today’s electrical BCIs, but they are only necessary for the highest-end applications. AI and neuroscience are converging because internal representations in models resemble representations in the brain, suggesting a shared information-processing framework. The long-term future of BCI extends beyond therapy toward augmentation, conscious-machine interfaces, and eventually brain-to-brain communication.
Data Points: People treated with early BCI vision therapy: More than 40 - Science announced more than 40 people had received one of its first BCI treatments restoring sight. Size of PRIMA implant: 2 mm x 2 mm - Tiny silicon chip implanted under the retina. Clinical trial scale: 17 sites - European clinical trial for the retinal prosthesis. Blindness prevalence: 200 million people globally - Age-related macular degeneration and related photoreceptor loss discussed as a major unmet need. Severe geographic atrophy prevalence: 1 million to a couple million - Subset of patients with severe age-related macular degeneration. Retina cell compression: 150 million rods and cones to 1.5 million optic nerve cells - Illustrates why stimulating downstream cells is harder and why preserving retinal computation matters. Retina intermediate layer: 100 million bipolar cells - Science targets bipolar cells to preserve image-like processing before compression. Motor decoding bandwidth: ~10 bits per second - Compared with keyboard/mouse and spoken language, used to argue current cortical motor BCIs are too limited for healthy users. Speech bandwidth: ~40 bits per second - Used as a rough comparison against cortical motor decoding. Typical typing bandwidth: ~20 bits per second - Compared with BCI motor decoding to show current devices are not yet advantageous for healthy users. Corpus callosum fibers: ~200 million fibers - Described as nature’s existing high-bandwidth bridge between hemispheres. Liver transplant perfusion usage: Over 75% in the US - Used to show how perfusion technology has become standard in transplantation. Potential near-term vision goal: Close to native acuity in 5-10 years - Hodak says this is possible though not current state. 2025/approval timing: Submit for approval later this year - Science expects regulatory submission for PRIMA after its clinical trial.
Pivotal Quotes: "I think it is very possible that the first people to live to a thousand are alive right now." — Max Hodak: Opening on longevity and the scale of near-term biotech change. "To me, it feels like we're firmly in the takeoff era now. Like something new has happened on Earth." — Max Hodak: Describing the pace and qualitative shift in biotech/BCI progress. "BCI we're going to come to see is not a specific product. I think there are going to be a bunch of BCI companies going after different applications where different types of probes will make sense." — Max Hodak: Explaining why BCI should be viewed as a category, not a single device.
Implications: Listeners should expect BCI to broaden from niche disability restoration into a multi-product industry spanning therapy, augmentation, and possibly consciousness research. The likely path is incremental at first, then rapidly transformative as better implants, biohybrids, and AI-neuroscience insights mature.
About Y Combinator Startup Podcast
We help founders make something people want. The Y Combinator Podcast is where builders talk about building. From the earliest days of an idea to scaling a company that changes the world, YC partners and founders share real stories, lessons, and tactics from the frontlines.