Bio-Computers: Brain Organoids as Processors

The line between biology and technology is blurring faster than many realize. While artificial intelligence dominates the headlines, a quieter revolution called “Organoid Intelligence” (OI) is taking place in wet labs around the world. Scientists are now successfully connecting lab-grown brain cells to computer hardware to create living, thinking biological processors.

What Are Brain Organoids?

To understand bio-computers, you first need to understand the hardware. These are not full human brains floating in jars. Brain organoids are tiny, three-dimensional clusters of neurons derived from human stem cells. They are usually about the size of a pen dot.

Researchers start with skin cells, reprogram them into stem cells, and then guide them to become brain cells. Over a period of months, these cells self-organize. They form complex structures and neural networks similar to a developing human brain. They can fire signals and communicate with one another.

Until recently, these organoids were used primarily to study diseases like Alzheimer’s or the effects of drugs. Now, researchers at institutions like Johns Hopkins University are turning them into biological hardware.

The Concept of Organoid Intelligence

The term “Organoid Intelligence” was popularized by Professor Thomas Hartung of Johns Hopkins. The goal is to leverage the unparalleled efficiency of the biological brain for computing tasks.

Silicon chips are fast at math, but biological brains are superior at learning and pattern recognition. Current supercomputers require massive amounts of energy to mimic what a human brain does on the energy equivalent of a sandwich.

The Energy Gap

• Frontier Supercomputer: This machine, housed in Tennessee, exceeds the computational speed of a single human brain. However, it requires 21 megawatts of energy and a massive cooling facility.
• Human Brain: A human brain operates on roughly 20 watts of power. It is incredibly efficient at processing uncertain data and making decisions.

By interfacing organoids with computers, scientists hope to build “biocomputers” that consume a fraction of the energy required by silicon chips while solving problems silicon struggles with.

Real-World Breakthroughs: DishBrain and FinalSpark

This is not theoretical. It is happening right now. Two major examples highlight how far this technology has come.

Cortical Labs and “DishBrain”

In 2022, an Australian startup called Cortical Labs made headlines with a system they nicknamed “DishBrain.” They grew roughly 800,000 human brain cells on top of a silicon chip containing a Multi-Electrode Array (MEA).

They taught this layer of cells to play the video game Pong. The computer sent electrical signals to the cells indicating where the ball was. If the cells moved the paddle correctly, they received a predictable, rhythmic electrical pulse (which neurons prefer). If they missed, they received chaotic, unpredictable noise.

To minimize the chaotic noise, the cells reorganized themselves and learned to play the game in under five minutes. This proved that lab-grown neurons could interact with a digital environment to achieve a goal.

FinalSpark’s Neuroplatform

In May 2024, a Swiss startup called FinalSpark launched the first online biocomputing platform. They call it the “Neuroplatform.”

This system allows researchers to access 16 human brain organoids remotely. It is essentially cloud computing, but the processors are biological. FinalSpark rents access to these neurons for roughly $500 a month. Their setup uses a closed-loop system where researchers can send electrical inputs and receive outputs, allowing them to train the organoids for AI tasks. This is the first time “wetware” has been commercially available as a service.

How the Interface Works

Connecting soft tissue to hard metal requires precise engineering. The core technology involves two main components.

1. Microelectrode Arrays (MEAs): The organoids sit on chips embedded with tiny electrodes. These electrodes act as a two-way radio. They record the electrical firing of the neurons and send electrical stimulation back to the cells.
2. Microfluidics: Biological processors need life support. A complex system of tiny tubes pumps a nutrient-rich fluid (media) over the cells to feed them and remove waste products. This is the “bloodstream” of the bio-computer.

Maintaining these systems is difficult. While a silicon chip can sit on a shelf for years, an organoid has a limited lifespan. FinalSpark, for example, reports that their organoids currently last around 100 days before degrading.

The Ethical Landscape

The rapid advancement of Organoid Intelligence raises immediate ethical questions. If we teach brain cells to compute, are we creating something that can feel?

Sentience vs. Intelligence

Scientists distinguish between intelligence (processing data) and sentience (having feelings or consciousness). Currently, these organoids are very primitive. They lack the sensory inputs and complex structure required for consciousness as we know it.

However, as these systems become larger and more complex, the line becomes blurry. Researchers are calling for “embodied intelligence” ethics guidelines. They need to ensure that as we scale up these biological processors, we do not inadvertently create entities capable of suffering.

The Future of Bio-Computing

We are unlikely to see biological processors in consumer laptops or phones anytime soon. Keeping the cells alive requires heavy machinery and strict temperature controls.

Instead, the future lies in hybrid data centers. We will likely see facilities where traditional silicon chips handle brute-force math, while racks of biological processing units handle complex learning and pattern recognition tasks. This hybrid approach could drastically reduce the carbon footprint of the AI industry.

Frequently Asked Questions

Are brain organoids conscious? No. Currently, they are simple clusters of cells. They lack the size, structure, and sensory input necessary for human-like consciousness or self-awareness.

Can a bio-computer replace my laptop? Not in the near future. Bio-computers require life-support systems to keep the cells alive. They are better suited for large data centers or research labs rather than personal devices.

Why use brain cells instead of AI? Brain cells learn much faster and use significantly less power. A human brain is millions of times more energy-efficient than a silicon chip when processing complex, unstructured information.

How long do the organoids live? In a biocomputing setup, they currently last a few months. For example, the organoids used by FinalSpark have a functional lifespan of roughly 100 days before they need to be replaced.

Is this technology available now? Yes, but mostly for researchers. Companies like FinalSpark allow universities and tech firms to rent time on biological processors through the cloud.