Futurist > Companies creating the future > 5 pioneering organizations building Synthetic Biological Intelligence
5 pioneering organizations building Synthetic Biological Intelligence: computers from living brain cells
Brains outperform computers by almost every meaningful measure: learning from minimal data, adapting continuously and running on far less power. For decades, this gap inspired silicon-based attempts to mimic the brain. Now pioneers within the growing Synthetic Biological Intelligence (SBI) sector are actually using neurons rather than trying to copy them.
SBI merges living biological neurons with computer hardware to perform genuine computational tasks. Neurons grown from human or animal stem cells and cultivated on multi-electrode arrays can learn, adapt, and process information, with potential advantages including dramatically lower energy consumption and greater learning capacity.
The number of organizations in the field is still limited, but the pace of progress in the last two years has been striking: the first commercial biological computers have reached the market, the first cloud-accessible neuron platforms have gone live, and fresh capital has arrived from defense agencies and leading venture firms. With more players likely to emerge, ethical questions raised by SBI remain, particularly around consciousness in engineered biological systems.
Below are five of the most significant players, ranging from commercial pioneers to the academic program that gave the field its name.
Cortical Labs
Coined the term “Synthetic Biological Intelligence” and became the first company to sell a working biological computer.
Stage: Early commercial
Total raised: ~$11M (2023 Series A led by Horizons Ventures, including In-Q-Tel)
Latest: CL1 commercially launched March 2025; cloud access (“Wetware-as-a-Service”) planned for H2 2025
The CL1 integrates human-derived neurons cultivated on a planar electrode array with silicon hardware, enabling a more flexible and learning-capable computing framework than fixed logic gates allow. The company’s foundational proof of concept came in 2022 when around 800,000 human and mouse neurons on a chip taught themselves to play Pong in real time.
The CL1 is positioned as a research tool with applications in drug discovery, clinical testing, and robotics, available at around $35,000 or through a cloud subscription model. Backers include the CIA’s venture arm In-Q-Tel, signaling early defense interest. No other company has a product this tangible on the market today.
FinalSpark
Running the world’s first commercially accessible biocomputing platform, built around living human brain organoids wired into the cloud.
Stage: Research platform / pre-commercial
Total raised: Undisclosed; seeking CHF 50M ($63M) for next stage
Latest: Neuroplatform operational since May 2024; 10-year roadmap presented in London, June 2025
The Neuroplatform comprises 16 brain organoids arranged in four multi-electrode arrays, each containing approximately 10,000 neurons, yielding roughly 160,000 neurons system-wide. The company uses dopamine rewards delivered via light-activated molecular cages to train the organoids, releasing dopamine through specific light frequencies when desired behavior occurs. The platform claims energy consumption a million times lower than conventional processors, a figure grounded in peer-reviewed work published in Frontiers in Artificial Intelligence.
Within ten years, FinalSpark aims to build bio-servers accessible via the cloud for generative AI workloads. Organoid lifespan has improved from a few hours to around 100 days, with a target of 200 days by 2026. Founded by two EPFL PhDs, the platform has run 24/7 and processed over 18 terabytes of neural data, making it one of the most substantive operating wetware research systems in existence.
Koniku
A Silicon Valley neurotechnology startup building “smell cyborgs”: wetware co-processors that fuse engineered biological neurons with silicon to detect explosives, drugs, and disease biomarkers in real time.
Stage: Development / pilot deployment
Total raised: ~$46.5M
Latest: Founding member of Oracle’s Defense Ecosystem in 2025; long-running Airbus partnership for aviation threat detection
The Koniku Kore fuses live neurons from mouse stem cells with silicon, creating a smell cyborg smaller than an iPhone and weighing under 600 grams. It engineers proteins in biological neurons to create precise receptor interactions that function as sensors and biological signal processors, capable of detecting volatile organic compounds associated with explosives, cancer, and drugs including fentanyl.
Founded by Nigerian neuroscientist Osh Agabi, who first proposed the concept at TEDGlobal in 2007, Koniku is the most application-focused of the wetware companies. Its use of mouse neurons sidesteps much of the ethical complexity around human cells, while its olfaction focus gives it a genuinely defensible niche: detecting smells remains one of biology’s hardest problems for silicon to replicate, and no algorithm comes close to a dog’s nose.
The Biological Computing Co. (TBC)
A San Francisco startup using lab-grown neurons as a co-processing layer to enhance frontier AI models, targeting computer vision and generative video with biological compute that thinks rather than just calculates.
Stage: Early commercial
Total raised: $25M seed (February 2026, led by Primary Ventures)
Latest: Commercial launch February 2026; flagship lab open in Mission Bay, San Francisco; first customers deploying generative models
TBC’s chips contain a multi-electrode array in which neurons, grown from stem cells reprogrammed into frontal cortex cells, are maintained at between 100,000 and 500,000 neurons per chip. Neural responses are decoded into richer representations and mapped onto foundation AI models via modular software adapters, enhancing model stability and efficiency without replacing the silicon stack beneath it.
Co-founded by neurosurgeon-scientists Alex Ksendzovsky and Jon Pomeraniec, and backed by Primary Ventures and Refactor Capital, TBC’s near-term commercial product is software: biological systems generate adapters once, and the deliverable to customers scales digitally. Over a longer horizon, the ambition is to put living neurons directly into the real-time inference loop, positioning biological computing as a genuine scaling path beyond brute-force silicon.
Johns Hopkins Organoid Intelligence Program
The research group that coined the term “organoid intelligence” and built the scientific and ethical foundations that every commercial player in this space is drawing on.
Stage: Academic research / pre-commercial
Funding: $15M NIH grant (March 2026, five-year); SURPASS internal funding; prior NSF support
Latest: Nature Communications Biology paper August 2025 demonstrating organoid learning and memory; DROIDp platform launched March 2026; whole-brain multi-region organoid grown July 2025
Scientists at Johns Hopkins have provided evidence that human brain organoids can replicate the fundamental building blocks of learning and memory, showing synaptic plasticity and increased expression of immediate early genes normally activated when the brain forms memories. The team has also created a prototype system for organoid intelligence experiments, with new methods for electrically stimulating and recording organoid activity, and novel machine learning tasks to test these systems.
The DROIDp platform uses human stem cell-derived brain organoids, advanced electrical sensors, and AI analytics to assess neural functions, with disease models including Alzheimer’s and SYNGAP1-related disorders. Led by Smirnova and Hartung, the program’s ultimate goal is growing organoids to one billion neural cells. No commercial spinout has emerged yet, but its embedded ethics framework and NIH backing make it the foundational infrastructure on which this sector depends.
