OBGNAUH OBGNAUH@OUTLOOK.COM

BIOLOGICAL COMPUTING · TECHNICAL DOSSIER

CL1
BIOLOGICAL COMPUTER

CL1 Biological Computer

When life begins to compute.

Explore the technology that fuses living neurons with silicon: from electrical spikes to programmable biological neural networks, and how a next-generation hybrid computing system actually works.

~45 min read · 12 modules · continuously updated · Last updated 2026-11-20

01 DEFINITION

What the CL1 actually is

The CL1 is a biological computer from Cortical Labs (Melbourne): human induced pluripotent stem cell (iPSC) derived neurons are cultured on a microelectrode array. Software stimulates them, reads their activity back through the same array, and closes the loop in real time.

It is not a general-purpose processor, nor a server that simply “uses cells for compute”. A more accurate description: a programmable neurophysiology platform that treats a living neural network as its computational medium. The unit still contains full life support, data acquisition electronics and a software stack.

Company and product background

Company
Cortical Labs, founded in Melbourne in 2019
Key result
DishBrain (2022, Neuron): closed-loop cultured neurons showed measurable task adaptation in a simulated world
Form factor
Rack-mounted unit integrating culture chamber, fluidics, thermal control, MEA interface and compute
Software
biOS (Biological Intelligence Operating System): world simulation, stimulation/acquisition and closed-loop scheduling
Commercial model
Unit sales plus cloud wetware-as-a-service
Public price
US$35,000 per unit in 2025; about US$20,000 each in 30-unit racks

Six facts to get straight first

Biological computing is often sold as a “mini-brain”. Before the engineering details, the boundaries matter.

01

It is not a complete human brain

Around 800,000 cultured neurons versus roughly 86 billion in a human brain — five orders of magnitude, with no brain regions, senses or body.

02

Neurons are not CPUs

Neurons have no instruction set, clock or deterministic addressing; they process information analogically through spike timing and synaptic weights.

03

Activity is not consciousness

Observable electrical activity, plasticity or even “learning” behaviour is not evidence of consciousness, sentience or self-awareness.

04

Electronics still do most of the work

Stimulation, amplification, sampling, denoising, decoding, experiment control and storage are all handled by conventional electronics and software.

05

Learning ≠ backpropagation

Biological networks rely on synaptic plasticity, homeostasis and network dynamics — no loss function, no gradient descent. The analogy is loose.

06

Public specs ≠ internal design

This dossier separates published specs, peer-reviewed findings, engineering estimates and undisclosed internals. Undisclosed parts are schematic only.

Timeline: from lab to product

  1. 2019

    Cortical Labs founded Official / public

    Founded in Melbourne, Australia, with the goal of turning living neurons into a programmable computing substrate.

  2. 2022

    DishBrain: neurons learn Pong Peer-reviewed

    A closed-loop system embodied cultured neurons in a simulated game world; performance improved within minutes. Published in Neuron.

  3. 2024

    CL1 prototype revealed Official / public

    Cortical Labs disclosed a rack-scale unit integrating neurons, life support and a compute interface.

  4. 2025

    CL1 goes on sale Official / public

    First units shipped from summer 2025 at a public price of US$35,000, alongside cloud wetware access.

Three kinds of “neural network”

DimensionBiological computer (CL1 class)Conventional computerArtificial neural network
Computational mediumLiving neurons + silicon interfaceSilicon transistorsFloating-point matrix math
Information carrierSpike timing, synaptic weights, field potentialsBinary voltage levelsNumeric tensors
ProgrammabilityShaped indirectly via stimulation, environment and reward — not instruction-levelFull instruction set, deterministicDefined by code and training
DeterminismLow: biological variability and noiseHighReproducible after training, near-deterministic on hardware
PowerTens of watts per unit (reported)Tens to hundreds of wattsTraining needs large GPU clusters
MaturityEarly research platform, limited volumeMature industryMature industry

Two terms you need

Wetware Computing

Wetware computing

An umbrella term for using biological tissue (cells, neural networks) as the computing or information-processing medium, as opposed to hardware and software. It stresses the biological substrate; it is not “biologically inspired algorithms”.

Synthetic Biological Intelligence

Synthetic biological intelligence

A term used by Cortical Labs for cultured neural networks that can be shaped through external interfaces and show adaptive behaviour in a closed loop. It blends engineering and marketing; its scientific boundaries are still debated.

Common misconceptions

“It is a human brain in a box”

No. It is a dispersed cultured network with no brain regions, senses, body or global organisation.

“The cells write code and run LLMs”

No. Tasks are defined by biOS and external software; neurons provide a shaped dynamical response.

“Biological computing has replaced GPUs”

Far from it. Current scale, stability and programmability cannot support general workloads.

“Buy one and culture neurons at home”

Human cell culture involves biosafety, ethics review and specialist facilities. A home setting is not compliant.

This is an independent technical dossier, not affiliated with Cortical Labs. CL1 is a product name; trademarks belong to their owners.

02 SYSTEM

Inside a biological computer

Five subsystems must work together: living medium, silicon interface, life support, electronics and software. If one fails, no usable data exists.

The medium itself: human iPSC-derived neurons that differentiate, mature and self-organise into a network.

Source and differentiation

  • iPSCSomatic cells are reprogrammed into induced pluripotent stem cells, then directed toward cortical-type neurons, using compliant donations.
  • Differentiation and maturationDifferentiation and maturation take weeks: neuronal markers appear, polarity establishes and synapses form.
  • Culture conditionsTemperature, pH, osmolarity, glucose, glutamine, neurotrophic factors and gas partial pressures jointly determine survival and phenotype.

Network and electrical activity

  • SynapsesExcitatory and inhibitory synapses shape the network; spontaneous activity often appears as synchronous bursts.
  • PlasticityShort-term plasticity, LTP/LTD and homeostatic regulation jointly change connection strength.
  • ReadoutExtracellular electrodes record field potentials generated by spikes, not intracellular membrane voltage.

Cultured neurons age, vary between batches and can be contaminated — a root cause of biological computing's instability.

Official / publicPeer-reviewedEngineering estimateUnverified

03 CLOSED LOOP

How it works: one full loop

Click any step for detail, or run the loop to watch one full stimulate–respond–decode cycle. Every waveform and number is simulated in your browser.

SIMULATED

04 TEARDOWN

Inside: interactive teardown

Below is a rotatable, zoomable, exploded schematic. It explains system layering and part relationships; it is not CL1's actual engineering drawing.

Drag to rotate · Scroll to zoom

Illustrative model — not the actual internal structure of CL1

05 MANUFACTURING

How a biological computer is built

Building one is not “putting chips and cells in a box”. Seven parallel engineering tracks converge into a single integration. Each phase below lists goals, work items and deliverables.

Lab-grade manufacturing ≠ electronics DIY

Any work with human cells requires biosafety level, ethics review, trained staff and compliant waste handling. This page explains principles, equipment needs and compliance only — never home culture protocols.

  1. GoalProve cells differentiate, attach and form a recordable network reliably.

    • Select cell source and compliant donor path (iPSC line or commercial bank)
    • Establish differentiation protocol and maturation timeline
    • Optimise medium, coating and feeding cadence
    • Validate neuronal phenotype with patch clamp, calcium imaging or MEA
    • Define viability and contamination assays

    Deliverable / gateReproducible culture protocol, baseline viability data and batch release criteria

    Key riskBatch variability, low differentiation efficiency, contamination

06 BILL OF MATERIALS

Materials and equipment database

Browse components by category: function, principle, specifications, estimated price, alternatives and public usage. Prices are public-channel estimates, not quotations.

Items without reliable quotes or specs are marked estimated or unverified; no vendor data is fabricated.

07 SOFTWARE

Software and programming

Programming a biological computer is not writing instructions; it is designing input encoding, closed-loop policy and decoders. Three kinds of code are separated below: real hardware control, pure simulation, and the in-browser demo on this page.

The following is an interface sketch, not the official Cortical Labs SDK. Real hardware must use vendor drivers, current limits and safety procedures.

python
# Closed-loop pseudocode (interface sketch, not an official SDK)
from cl1_sdk import Device, MEA, LifeSupport   # hypothetical interface

dev = Device.open("cl1-01")
mea = MEA(dev, channels=64, sample_rate_hz=25_000)
life = LifeSupport(dev)

with life.maintain(temperature_c=37.0, co2_pct=5.0, ph_range=(7.25, 7.40)):
    mea.check_impedance(max_kohm=500)
    for trial in range(120):
        ball = world.next_state()               # 1. world state
        stim = encode_rate(ball, n_channels=16) # 2. state -> stimulation
        mea.stimulate(stim, amplitude_ua=8, biphasic=True)  # 3. inject charge
        spikes = mea.record(duration_ms=50)     # 4/5. record activity
        action = decode_population(spikes)      # 6/7. decode to action
        world.apply(action)                     # 8. feedback to world
        if score(action, ball) > 0:
            apply_reward(mea, pattern=stim)     # 9. reward shaping
# Real hardware must use vendor current limits and safety procedures.

08 LABORATORY

The biological computing laboratory

Behind one device sits an entire facility and governance layer. Functional zones, staffing, qualifications and compliance are outlined below.

Functional zones

Cell culture lab

Controlled clean environment, biosafety cabinet, CO₂ incubator, inverted microscope, centrifuge.

Engineering estimate

Biosafety facilities

BSL level per local regulation; negative pressure, HEPA, autoclave and compliant waste disposal.

Official / public

Chip R&D area

Lithography/sputtering/bonding can be outsourced; in-house impedance testing, probe station and clean bench.

Engineering estimate

Electronics bench

Oscilloscope, source meter, low-noise supplies, pre-compliance EMI and rework stations.

Engineering estimate

Control centre

Host computers, storage, remote access and scheduling; network and permission isolation.

Engineering estimate

Data acquisition

High-bandwidth storage, time sync, backup and integrity checks.

Engineering estimate

Maintenance area

Spares, calibration tools, cleaning and sterilisation flow, physically separated from culture.

Engineering estimate

People and qualifications

  • Cell biologistAseptic technique, culture, differentiation and contamination troubleshooting.Official / public
  • Hardware engineerAnalog front end, low-noise design, EMC and embedded timing.Engineering estimate
  • Software / algorithmsReal-time systems, signal processing, decoding and visualisation.Engineering estimate
  • Biosafety officerRisk assessment, training, incident response and audits.Official / public
  • Ethics and data complianceDonor consent, data governance and cross-border transfer review.Official / public

Compliance checklist

  • Lawful cell sourcing, informed consent and ethics approval
  • Biosafety level aligned with local regulations
  • Training, health monitoring and incident plans
  • Compliant waste, effluent and sharps disposal
  • Data privacy: de-identification and access auditing
  • Material transfer agreements and export controls for collaboration

Exact levels and approvals vary by jurisdiction; follow local regulators and institutional rules.

09 INDEPENDENT R&D

Independent R&D: a modular biological computing testbed

The goal is not to clone the CL1, but to build a modular testbed that validates the principles. Three tiers progress step by step, each valid on its own.

A

A · Software simulation

Entry / no biological material

Build a spiking network with LIF or Izhikevich neurons, simulating encoding, closed-loop decoding and STDP.

Technical requirements

  • Understand spiking networks and STDP
  • Design encoders and decoders
  • Reproducible experiment logs

Hardware

  • Laptop or desktop
  • Optional: SBC for better real-time behaviour

Software

  • Python + NumPy/SciPy
  • Brian2 or NEST
  • Matplotlib or web visualisation

Hard problems

  • Keeping results interpretable, not overfit
  • Loop latency and numerical stability

Budget range¥0 – US$450

Timeline1 – 3 weeks

Success criteria

  • Task metric improves over time
  • Performance drops when plasticity is disabled (ablation)
  • Full logs reproduce the run

This tier involves no living cells and must not be marketed as a real biological computer.

B

B · Electronic testbed

Intermediate / interface & life-support emulation

Build “electronic neurons” in analog circuitry, emulate MEA stimulation and acquisition with multichannel DAC/ADC, and monitor temperature, humidity and fluid level.

Technical requirements

  • Analog circuits and low-noise layout
  • Multichannel synchronous acquisition
  • Basic control theory and PID

Hardware

  • Multichannel DAC/ADC board
  • Op-amps, filters and shielded enclosure
  • Sensors and actuators for temperature, humidity, level

Software

  • C/C++ or Python real-time acquisition
  • LabVIEW, Qt or a web panel
  • Data logging and replay

Hard problems

  • Pushing noise down to microvolts
  • Crosstalk and common-mode rejection
  • Artefacts and stimulation blanking

Budget rangeUS$4,000 – US$21,000

Timeline3 – 6 months

Success criteria

  • Stable acquisition on 16+ channels
  • Injected pulses are detected
  • Out-of-range environment triggers protection

This tier involves no living cells and must not be marketed as a real biological computer.

Budgets are engineering estimate ranges and vary widely with region, FX, used equipment and in-house work.

10 APPLICATIONS

Applications and maturity

Separating validated research uses from long-range speculation is essential to judging where this technology really stands.

Official / publicValidated research use

Neuroscience and learning

Study plasticity, network dynamics and behaviour in controlled loops.

Peer-reviewedEarly exploration

Drug and toxicity screening

Observe drug effects on human neural activity patterns, complementing animal models.

Peer-reviewedEarly exploration

Disease modelling

Patient-derived iPSC networks for electrophysiological phenotypes in epilepsy and neurodegeneration.

Peer-reviewedEarly exploration

Biological computing

Explore encoding, decoding and energy limits of living networks as a computational medium.

Peer-reviewedEarly exploration

Closed-loop robotics

Map neural responses to control signals and study adaptation in embodied systems.

Peer-reviewedEarly exploration

Low-power computing research

Compare energy scaling of biological and silicon approaches on narrow tasks.

UnverifiedLong-range speculation

AI architecture inspiration

Derive sparse, event-driven computing ideas from biological dynamics.

UnverifiedLong-range speculation

General-purpose compute replacement

Not feasible today: scale, stability and programmability fall far short.

UnverifiedLong-range speculation

Consciousness or AGI

No scientific evidence; it should not be an engineering claim.

11 COST

Cost structure and hard problems

The calculator below lets you edit quantities and unit prices, totals live, and switches between CNY and USD. Prices are midpoints of publicly visible ranges — engineering estimates, not quotations.

Preset
Stage Item Qty Unit price (editable) Subtotal Basis
Grand total ¥0 $0

Items without formal quotes are inferred from public retail prices or comparable products; error may reach ±50%.

Six genuinely hard problems

Long-term sterility Official / public

Over weeks or months, one connector, bubble or handling error can invalidate an entire run.

Signal quality Peer-reviewed

Microvolt signals coexist with millivolt stimulation; artefact blanking and noise control decide how much data is usable.

Biological variability Peer-reviewed

Batch, passage and history change network dynamics; designs need controls and replication.

Reproducibility Peer-reviewed

The same stimulation may yield different responses across cultures; cross-lab replication remains a core challenge.

Device lifetime Official / public

Cultures last months, so core biological components and consumables must be replaced periodically.

Compliance and ethics Official / public

Human cells, data governance and cross-border material transfer add institutional and time cost.

12 LIBRARY

Technical library

Filter by type, then open an entry for its abstract and source link. Links point to official pages, journal DOIs or vendor documentation.

13 GLOSSARY

Glossary

Search or browse by category. Each entry gives a concise definition and where it appears.

  • Wetware Computing

    Concept Peer-reviewed

    Computing that uses biological tissue as its medium.

  • Synthetic Biological Intelligence (SBI)

    Concept Official / public

    Cortical Labs' term for adaptive behaviour shown by cultured neural networks.

  • iPSC

    Biology Peer-reviewed

    Induced pluripotent stem cells, reprogrammed from somatic cells and differentiated into neurons.

  • MEA

    Interface Official / public

    Microelectrode array: a grid of electrodes for recording and stimulating electrogenic cells in vitro.

  • biOS

    Software Official / public

    Cortical Labs' Biological Intelligence Operating System.

  • LFP

    Signal Peer-reviewed

    Local field potential: low-frequency signal reflecting population synaptic activity near an electrode.

  • Spike

    Signal Peer-reviewed

    The extracellular signature of a neuronal action potential, a brief high-frequency event.

  • Burst

    Signal Peer-reviewed

    A cluster of synchronous spikes across neurons in a short window.

  • STDP

    Learning Peer-reviewed

    Spike-timing-dependent plasticity: synaptic strength changes with pre/post spike timing.

  • LTP / LTD

    Learning Peer-reviewed

    Long-term potentiation / depression: persistent changes in synaptic strength.

  • Closed loop

    System Official / public

    Stimulation, response, decoding and re-stimulation form a real-time loop.

  • Latency

    System Engineering estimate

    Delay from acquisition to feedback; it determines whether a loop is viable.

  • Charge balance

    Engineering Peer-reviewed

    Equal positive and negative charge per stimulus pulse, preventing electrode polarisation and cell damage.

  • Impedance

    Engineering Peer-reviewed

    AC resistance at the electrode–electrolyte interface; affects noise and stimulation efficiency.

  • Artefact

    Engineering Peer-reviewed

    Interference on recording channels during stimulation; removed by blanking or template subtraction.

  • Biosafety level

    Compliance Official / public

    Laboratory containment levels classified by biological risk.

  • MTA

    Compliance Official / public

    Material Transfer Agreement governing biological material between institutions.

  • LIF neuron

    Model Peer-reviewed

    Leaky integrate-and-fire, one of the most common simplified neuron models.

  • Organoid

    Biology Peer-reviewed

    A three-dimensional miniature tissue self-organised from stem cells.

  • WaaS

    Business Official / public

    Wetware-as-a-Service: remote cloud access to biological computing resources.

Frequently asked questions

Is the CL1 an actual human brain?

No. It contains roughly 800,000 cultured neurons with no brain regions, sensory input or body, and no evidence of consciousness.

Can it replace GPUs for AI training?

No. Current scale, stability and programmability cannot support general training; it targets neuroscience research and closed-loop experiments.

How long do the cells live, and what maintenance is needed?

Cortical Labs cites up to about six months with internal life support. Maintenance includes medium exchange, temperature and gas control, contamination monitoring, impedance tracking and calibration.

Can an individual culture neurons at home?

Neither advisable nor compliant. Human cells require biosafety, ethics review and specialist facilities. Individuals should start with tier A simulation or tier B electronics.

Are the waveforms and numbers real experimental data?

No. Every animated trace is simulated in the browser to explain mechanisms. Real data must come from controlled experiments and publications.

Does learning a task imply intelligence?

No. Task adaptation can be explained by low-level plasticity and network dynamics; it does not imply general intelligence or consciousness.

Official / publicPeer-reviewedEngineering estimateUnverified