Yes — there is extensive and rapidly accelerating evidence of serious attempts to grow living human brain cells and use them as a computational substrate for AI. This field has been formally named Organoid Intelligence (OI) and has moved from laboratory curiosities to commercially available products and government-funded research programs within just a few years.
The Core Concept
Organoid Intelligence uses lab-grown brain organoids — miniature, self-organizing three-dimensional neural tissues derived from human stem cells — as biological hardware for computing . Unlike traditional silicon-based AI, which merely models neural behavior, OI systems compute via the actual electrical activity of living neurons and glia, communicating through spikes and complex network dynamics . The biological neurons naturally exhibit synaptic plasticity (long-term potentiation and depression), parallel processing, and emergent behavior not explicitly programmed at design time .
The key advantage driving this research is energy efficiency. The human brain operates on roughly 20 watts, while training a state-of-the-art AI model can consume megawatts . FinalSpark claims its bioprocessors consume a million times less power than traditional digital processors .
Key Companies and Projects
Cortical Labs (Australia)
Cortical Labs gained worldwide attention in 2022 with DishBrain, an experiment where approximately 800,000 lab-grown human and mouse neurons on a chip learned to play Pong within about five minutes . The neurons received unpredictable electrical signals when they "got it wrong," compelling them to adapt — a test of Karl Friston's Free Energy Principle .
Building on this, Cortical Labs launched the CL1 in 2025 — described as the world's first commercially available biological computer . Priced at $35,000 per unit, the CL1 houses 800,000 human brain cells cultivated in a nutrient-rich solution on a silicon chip . It runs on a Biological Intelligence Operating System (biOS) that creates a simulated world for the neurons, sending information directly to them and reading their impulse responses in real time . Neurons can survive up to six months before needing replacement . The company also offers cloud-based access through what it calls "Wetware-as-a-Service" (WaaS) .
FinalSpark (Switzerland)
Swiss startup FinalSpark launched its Neuroplatform — the first publicly rentable biocomputer platform — available to researchers worldwide for approximately $500 per month . The platform hosts brain organoids of about 10,000 living neurons each (0.5 mm in diameter), grown from stem cells, kept alive in incubators at body temperature, and wired into electrical circuits via multi-electrode arrays . The system uses dopamine as a reward signal to train neurons, mimicking the brain's natural learning mechanism . Over three dozen universities have expressed interest, and nine institutions have already received free research access .
Koniku (United States)
California-based Koniku takes a different approach, engineering biological neurons with olfactory receptors to create chips that can detect and classify smells . Their Konikore device, unveiled at TEDGlobal 2017, combines living neurons with silicon for applications in explosives detection, healthcare diagnostics, and industrial monitoring . Koniku has partnered with Airbus for aviation threat detection and in 2025 became a founding member of Oracle's Defense Ecosystem .
Government Investment and Academic Programs
The field has attracted serious government backing. The U.S. National Science Foundation invested $14 million in seven interdisciplinary research projects through its "Biocomputing through EnGINeering Organoid Intelligence" (BEGIN OI) program . Each project received $2 million, and uniquely, the NSF required an ethicist as co-principal investigator on every proposal, with ethics evaluated on equal footing with the science . DARPA has also invested in organoid-based biocomputing .
A flagship project at Johns Hopkins University, led by Thomas Hartung (a key figure behind the 2023 Baltimore Declaration), seeks to incorporate reward-based learning into brain organoids by connecting them to electronic shell arrays and using chemical signals to teach them to play video games and guide small robots . A team at UC San Diego has proposed using organoid-based systems to predict oil spill trajectories in the Amazon by 2028 .
The Baltimore Declaration
In 2023, a consortium of researchers published the Baltimore Declaration, formally establishing Organoid Intelligence as a field . The declaration calls on the scientific community to explore brain organoid cell cultures for advancing brain understanding and developing biological computing, while proactively addressing ethical concerns . It urges breakthroughs in stem cell technology, bioengineering, brain-machine interfaces, and machine learning — while remaining vigilant about the possibility that organoids could develop forms of consciousness .
Current Capabilities and Limitations
The technology remains in its infancy. Current biocomputers can:
However, significant limitations persist:
Lifespan: Neurons survive only weeks to months before dying — the CL1's neurons last up to six months , while FinalSpark organoids require regular replacement .
Scale: Current organoids contain thousands to hundreds of thousands of neurons, compared to the human brain's ~86 billion .
Reproducibility: Brain organoids self-organize unpredictably, making consistent results difficult .
Capability gap: These systems display only simple adaptive responses, not anything resembling higher cognition or intelligence .
Interfacing: Reading from and writing to biological neurons with precision remains a major engineering challenge .
Ethical Concerns
The use of human brain tissue as computing hardware raises profound ethical questions . A November 2025 gathering at the Asilomar conference center brought together researchers, ethicists, and legal experts to discuss governance, with much conversation focused on how to determine whether lab-cultured neural constructs have developed sentience or consciousness . Leading organoid researchers in the biomedical space have expressed concern that the push into biocomputing could trigger a public backlash against all organoid research, including vital disease-modeling work .
Key ethical issues include whether organoids could attain consciousness, the moral status of such constructs, donor consent and genetic privacy, and whether existing legal frameworks for human or animal research apply . Most experts agree that current organoids are not conscious, but as they grow more complex, these questions will become increasingly urgent .
The Bigger Picture
This research sits at a fascinating intersection of your interests in quantum physics, consciousness, and AI. The fundamental question driving it — whether biological neurons can provide a form of "actual intelligence" rather than the statistical pattern-matching of current AI — echoes deeper philosophical questions about the nature of mind and computation. As Cortical Labs' chief scientist Brett Kagan put it: "The only ground truth we had for true generalized intelligence were biological brain cells" . Whether harnessing those cells in a dish can ever bridge the gap between silicon-based pattern recognition and genuine understanding remains one of the most provocative open questions at the frontier of both AI and neuroscience.
References
Organoid Intelligence: Biohybrid Neural Computing - Emergent Mind - Organoid Intelligence uses living brain organoids and virtual models to drive adaptive computation, ...
Exclusive Look at CL1: One-on-One w/ Cortical Labs' Chief ... - Fusing neurons with silicon chips might sound like science fiction, but for Cortical Labs, it repres...
Living brain-cell biocomputers are now training on dopamine - Current AI training methods burn colossal amounts of energy to learn, but the human brain sips just ...
These Living Computers Are Made from Human Neurons - In the search for less energy-hungry artificial intelligence, some scientists are exploring living c...
FinalSpark's Neuroplatform: the era of organic computing has begun! - The era of organic computing has begun! Swiss startup FinalSpark has just launched its Neuroplatform...
Cortical Labs - CL1 - The world's first code deployable biological computer. Test how lab-grown neurons process informatio...
FinalSpark Launches the First Remote Research Platform ... - FinalSpark, a pioneering Swiss biocomputing startup, has launched the first-ever online platform ena...
Cortical Labs Launches $35K Biological Computer Built on ... - Cortical Labs has launched the CL1, a biological computing system that integrates human brain cells ...
Brain organoids and organoid intelligence from ethical, legal, and ... - Human brain organoids, aka cerebral organoids or earlier “mini-brains”, are 3D cellular models that ...
Brain organoid scientists worried by push into biocomputing - Despite widespread skepticism, the National Science Foundation and DARPA have invested millions of d...
The Baltimore declaration toward the exploration of organoid ... - by T Hartung · 2023 · Cited by 29 — TH is employed by, and inventor on a patent by Johns Hopkins. Un...
Beyond consciousness: Ethical, legal, and social issues in ... - by M Kataoka · 2024 · Cited by 19 — This study aims to provide a comprehensive review of the ethical...
The Baltimore declaration toward the exploration of ... - by T Hartung · 2023 · Cited by 29 — This is part of an article hub The Baltimore declaration toward ...
Why Lab-Grown Brain Cells Might Never Become Conscious - Philosopher Matthew Owen questions whether tiny brain cell clusters called organoids could ever be s...
TEDGlobal: The computer that can smell explosives - Koniku Kore is an amalgam of living neurons and silicon, with olfactory capabilities — basically sen...
NSF invests $14M in bioengineered systems and ethical ... - The U.S. National Science Foundation has invested $14 million in seven interdisciplinary research pr...
Cortical Labs Launches $35K Biological Computer Built on Human Brain Cells
Hardware, Software, Meet Wetware: A Computer With 800,000 Human Neurons - The Cortical Labs CL1 costs $35,000 and has 800,000 human brain cells living and growing in a nutrie...
How Scientists Are Growing Computers From Human Brain ... - The coming years will determine whether organoid intelligence transforms computing or becomes a shor...
Australian startup Cortical Labs unveils 'world’s first' commercial biological computer - Company says its Synthetic Biological Intelligence is a “more advanced and sustainable form of AI”
Grant 2515214 The Johns Hopkins University - Project Grant 2515214 worth $2.0M was awarded to The Johns Hopkins University on 8/1/25 by the Emerg...
Powered by Human Brain Cells: Cortical Labs CL ... - Computers powered by human brain cells are officially entering the market.
No comments:
Post a Comment