This Startup Fused Human Brain Cells with Silicon Chips | E2295

Key Concepts

• Biological Computing: The integration of human neurons with silicon chips to create a hybrid computing platform.
• CL1: A rack-mountable biological computer unit developed by Cortical Labs, featuring a neural chamber, life support systems, and silicon-based I/O.
• Biological Data Center: A facility housing multiple CL1 units, utilizing on-site stem cell cultivation to maintain neural compute capacity.
• Sample Efficiency: A measure of how quickly a system learns; biological neurons have demonstrated 5,000x greater sample efficiency in reinforcement learning compared to traditional GPU-based systems.
• Microfluidics (PDMS): A technology using polydimethylsiloxane (PDMS) to create channels that constrain and direct neural growth, allowing for circuit-like control of neural communication.
• Group 4 UAS: Large unmanned aerial systems (over 1,320 lbs) used for commercial and defense logistics.
• Hybrid Propulsion: An aircraft architecture combining turbo-diesel engines for range and electric motors for high-power, quiet takeoff and landing.

---

1. Cortical Labs: Biological Computing

Cortical Labs is pioneering the fusion of biological neurons with silicon chips. Their flagship device, the CL1, acts as a "biological computer" that uses human neurons (grown from stem cells) to perform computational tasks.

• Hardware Architecture: The CL1 is a 3U rack-mountable unit. It includes a neural chamber maintained at 37°C, with integrated life support systems mimicking biological organs: pumps (heart), filtration units (kidneys), and gas mixers (lungs).
• Performance: While a human brain contains ~100 billion neurons, the CL1 utilizes ~200,000 neurons, which is sufficient for specific reinforcement learning tasks.
• Key Advantage: In reinforcement learning benchmarks, these biological systems proved 5,000 times more sample efficient than GPU-based systems.
• Ethical Framework: The company maintains a strict "red line" against creating conscious systems, citing the ethical imperative to prevent suffering. They engage proactively with bioethicists and religious institutions (e.g., the Vatican) to ensure alignment with ethical standards.
• Commercial Strategy: Rather than just selling hardware, they are building a "Biological Data Center" in Melbourne and Singapore to democratize access via the cloud, allowing developers to program these systems using Python and standard SDKs.

2. Pyka: Large-Scale Autonomous Drones

Pyka focuses on large-scale, autonomous electric and hybrid aircraft for agriculture and logistics.

• Agricultural Application: Their "Pelican" drone is currently deployed at scale in Brazil, where it performs crop spraying. It is significantly more efficient than traditional 8,000 lb "air tractor" planes, burning ~2 gallons of fuel per hour compared to 55 gallons.
• Drop Ship (Logistics): A second-generation cargo drone designed for defense and commercial logistics. It features a hybrid turbo-diesel/electric architecture:
  • Diesel: Used for long-range cruise (up to 1,000 miles).
  • Electric: Used for high-power, ballistic takeoff and landing (under 600 ft) and quiet operation.
• Vertical Integration: To maintain quality and reliability, Pyka designs its own motors, batteries, avionics, and airframes. This vertical integration allows them to create "NDA-compliant" (non-Chinese supply chain) versions of their hardware.
• Regulatory Challenges: The primary barrier to domestic US adoption is not technology, but FAA regulations regarding "Beyond Visual Line of Sight" (BVLOS) operations. Pyka is currently working with regulators to expand these operational limits.

---

3. Step-by-Step Processes & Methodologies

• Neural Maintenance: Neurons are kept alive via a nutrient-rich fluid system. The primary maintenance task involves replacing filtration cartridges (kidneys) every 4–6 months to prevent protein buildup.
• Reinforcement Learning (Doom Example): Developers use "auditory stimulus" (disordered vs. ordered inputs) as reward/punishment signals. The system learns to optimize gameplay (e.g., not wasting ammo) through these feedback loops.
• Hardware Development (Pyka): Pyka utilizes a rapid iteration cycle, moving from CAD design to first flight in as little as 180 days. They prioritize "product-reliability fit" by deploying units to remote, high-intensity environments (e.g., Brazilian farms) to gather real-world data for continuous retrofitting.

---

4. Notable Quotes

• Han (Cortical Labs): "You do not want to create conscious systems because ethically a conscious system has the ability to suffer and we do not want any suffering to come about from any technology."
• Michael (Pyka): "The thing that people don't understand is... the cost of the chemical that is being sprayed is typically about four times the cost of the application. So, the Pelican has a bunch of things about it that make it a massively superior solution for actually applying the chemical."

---

5. Synthesis and Conclusion

Both companies represent a shift toward specialized, high-efficiency hardware. Cortical Labs is challenging the "bitter lesson" of AI (which relies on throwing massive compute at problems) by utilizing the inherent efficiency of biological intelligence. Pyka is challenging traditional aviation by proving that autonomous, vertically integrated hardware can outperform legacy machinery in both cost and environmental impact. Both founders emphasize that the path to success in hardware is not just building the device, but owning the full stack—software, hardware, and the data loop—to ensure a seamless, reliable experience for the end user.