Premium Biotech Journal — Declassified Technical CODE/AURA 2026 SUBSTRATE · P(Meaning | ε)
REF: NEURAL-DOOM-V1.0 · TIME: T=001 / 20

The ultimate hardware test
becomes a biological milestone

Neurons living on a silicon chip - Cortical Labs
CORTICAL LABS · ACTUAL NEURONS ON SILICON
Lab-grown human neurons on a silicon chip — Cortical Labs © 2025
For decades, the ultimate tech meme has asked: "Can it run Doom?"
Cortical Labs CL1 Biocomputer
CL1 BIOLOGICAL COMPUTER
The CL1 — World's first code-deployable biological computer
Now, an Australian biotech company has trained 200,000 living human brain cells to play the game.
Premium Biotech JournalNotebookLM × CodeAura 2026
REF: NEURAL-TEST-V1.4 · TIME: T=002 / 20

200,000 living neurons navigate
a 3D digital environment

Neurons on MEA chip - Cortical Labs DishBrain
200,000 HUMAN NEURONS · LIVE
Cortical cells on high-density multielectrode array (HD-MEA) — CC BY Cortical Labs
IN-GAME ACTION
The biological player wandered 3D corridors, encountered enemies, and fired weapons.
HARDWARE LINK
Cells successfully hooked into the game via a silicon chip. Neurons receive spatial and visual data as electrical stimulation patterns.
NEURON CONTROL ACTIVE — BIO-FEEDBACK SIGNAL
OBSERVED BEHAVIOR
  • Played clumsily and died often
  • Demonstrated "adaptive, real-time goal-directed learning"
Premium Biotech JournalDeclassified Technical · NotebookLM
SOURCE PROCESS · CORTICAL LABS03 / 20

Cortical Labs is merging
biology with silicon

STEP 1: THE SOURCE
Stem cells
Led by Brett Kagan, the team takes a small bit of blood or skin — never extracting from human brains.
NON-INVASIVE COLLECTION
STEP 2: THE CONVERSION
🧬⚗️
Specific cells isolated and converted into induced pluripotent stem cells (iPSCs) in a laboratory environment.
iPSCs
STEP 3: THE SUPPLY
Neurons on chip
Stem cells generate an indefinite supply of neural cells for hybrid organic computing.
SCALABLE NEURAL COMPUTING
Premium Biotech JournalNotebookLM × CodeAura 2026
BIOLOGICAL COMPUTING EVOLUTION04 / 20

From 2D logic to 3D navigation:
a massive computational leap

2021: DISHBRAIN
PONG
  • Trained early biocomputer to play Pong
  • Navigating 2D data took over 18 months
2024–2026: THE DOOM BENCHMARK
DOOM
  • Rapid transition to a complex 3D shooter
  • Achieved in approx. one week by one developer
  • Signals biological computing moving from crawl to sprint
2D LOGIC → 3D NAVIGATION — DIMENSIONAL SHIFT
Premium Biotech JournalDeclassified Technical · NotebookLM
REF: CL1-HARDWARE-V1.005 / 20

Inside the CL1: The first
code-deployable biological computer

CL1 Biological Computer device - Cortical Labs
CL1 BIOCOMPUTER · CORTICAL LABS · 2025
CL1 — official product image © Cortical Labs 2025
NEURAL LAYER
800,000
Living human neurons per unit
FLUID CONTAINMENT · LIFE SUPPORT
Self-contained life-support system. Sustains biological units for up to 6 months. Pumps, gas mixing, temperature control, filtration.
SILICON BASE + biOS
Electrodes integrate standard Python code directly into living tissue. 59 inputs. Sub-millisecond latency.
PRICE: ~$35,000 USD · SHIPPING: 2026
Premium Biotech JournalDeclassified Technical · NotebookLM
BIOCOMPUTING WORKFLOW · 16:906 / 20

The Biocomputing Workflow bridges
digital code and living tissue

1. GAME ENGINE DATA
Doom outputs 3D spatial and visual data through the game engine in real time.
4. NEURON PROCESSING
Cells react to the electrical environment and send pulses back to the game to act.
2. PYTHON INTERFACE (biOS)
Digital code translates game states into biological instructions via standard Python API.
3. ELECTRICAL PULSES
Visual data converted into specific electrical stimulation patterns delivered to neurons.
CL1 Cortical Labs biological computer - working
CL1 IN OPERATION
CL1 running live neural interaction — © Cortical Labs 2025
Premium Biotech JournalNotebookLM × CodeAura 2026
DIGITAL REALM / BIOLOGICAL REALM07 / 20

A standard Python interface removes
the biology degree barrier

● ● ●
import biOS
send_stimulation(pulse)
read_neural_output()
train_epoch(doom_env)
DIGITAL REALM — biOS PYTHON API
  • Independent developer Sean Cole achieved this integration in about one week with no biological computing experience
  • By allowing anyone to use standard Python, Cortical Labs created a "biOS" — removing the biology PhD barrier forever
BIOLOGICAL REALM — LIVING NEURONS ON ELECTRODE ARRAY
Premium Biotech JournalDeclassified Technical · NotebookLM
SHARED LANGUAGE · 16:908 / 20

Biology and silicon communicate through
the shared language of electricity

🎮
DIGITAL INPUT
The cells lack optical input. They cannot "see" a screen directly.
THE TRANSLATION
Engineers mapped digital visual data into precise spatial electrical patterns delivered via 59 electrodes.
Neurons receiving signals
BIOLOGICAL RECEIPT
Active brain cells generate electrical pulses naturally. System delivers targeted pulses in their native language.
Premium Biotech JournalNotebookLM × CodeAura 2026
SENSORY INPUTS (CHAOS) → ORDER / PREDICTION09 / 20

The Free Energy Principle forces
the neurons to adapt and play

NO INNER DESIRE
The cells don't "like" playing Doom. The temptation to anthropomorphize is a mistake.
THE BIOLOGICAL MANDATE
They are driven by the Free Energy Principle: neural systems act to predict their environment and minimize chaos.
MASTERY AS SURVIVAL
The cells learn to play because mastering the environment makes electrical inputs predictable and safe.
ADAPTATION SIGNAL — CHAOS → ORDER
SUBSTRATE Research · P(Meaning | ε) ∝ Tr(ρ · Π_Data)CodeAura 2026
PREDICTABLE = SURVIVAL · CHAOTIC = IMPOSSIBLE10 / 20

Neural systems adapt because an
unpredictable world is impossible to live in

PREDICTABLE = SURVIVAL
🤚 → 🥤
Reach for a drink. It's a drink. Outcome predicted. World is liveable. Cells stabilize.
CHAOTIC = IMPOSSIBLE
🤚 → 🐔💥
Reach for a drink. It turns into a chicken and a firework. Chaos. That world is impossible.

"If I reach for an empty can of drink and I successfully predict the outcomes of my actions, that's a world I can live in. But if it sometimes turns into a chicken and a firework, that world would be impossible to live in."

— BRETT KAGAN, CHIEF SCIENTIFIC OFFICER · CORTICAL LABS
Premium Biotech JournalDeclassified Technical · NotebookLM
COMPARATIVE LEARNING TRAJECTORIES · DECLASSIFIED TECHNICAL11 / 20

Biological hardware demonstrates adaptive,
real-time goal-directed learning

COMPARATIVE LEARNING TRAJECTORIES
Time Biological Silicon ML Adaptation
KEY INSIGHTS
  • While the system plays clumsily and loses often, its initial learning phase is profoundly different from AI
  • The biocomputer reached its performance level faster than traditional silicon-based machine learning systems — Cortical Labs
Premium Biotech Journal · Declassified TechnicalNotebookLM × CodeAura 2026
PREMIUM BIOTECH JOURNAL12 / 20

Silicon AI's massive appetite for electricity
creates a power wall

THE SILICON POWER WALL
🖥️⚡🔥
  • Stakes extend far beyond video games
  • Traditional AI requires astronomical electricity to train and run
MASSIVE ENERGY CONSUMPTION
VS
BIOLOGICAL EFFICIENCY
Low-power biological computing
  • Neurons perform complex calculations far more efficiently
  • CL1 powered by less energy than a standard calculator
~0.000001 WATTS BIOLOGICAL LAYER
Premium Biotech JournalDeclassified Technical · NotebookLM
ROADMAP · LOW-POWER BIOLOGICAL DATA CENTERS13 / 20

The roadmap leads directly to
biological data centers

Melbourne Singapore
THE VISION
  • Replace massive, power-hungry server farms with hyper-dense biological processing units
THE REALITY — BLOOMBERG, MARCH 2026
  • Neurons survive up to six months on automated life support — Cortical Labs already planning the world's first commercial biological data centers in Melbourne and Singapore
Premium Biotech JournalDeclassified Technical · NotebookLM
REF: NEURAL-TEST-V1.4 · MEDICAL APPLICATIONS14 / 20

Active neural units offer unprecedented
ways to test neurological drugs

Cultivation
Stem cells grown into a functioning neural chip — a living, active computational substrate ready for testing.
Activation
The chip is hooked into a simulated task (like a game or puzzle) to establish active cognitive baselines.
Intervention
Experimental drugs introduced to living cells while they compute — scientists monitor real-time cognitive impact.
Living neurons for drug testing on chip
ACTIVE NEURAL CULTURE · DRUG TESTING PLATFORM
Living human neurons as drug testing substrate — © Cortical Labs 2025
Bypassing animal models by studying active, computing human tissue offers an entirely new approach to drug development
REF: NEURAL-TEST-V1.4 · ACTIVITY LEVEL: HIGH SPIKENotebookLM × CodeAura 2026
INTELLIGENCE THRESHOLD · CONSCIOUSNESS SCALE15 / 20

The Intelligence Threshold shifts our
understanding of scale

🪱
C. elegans worm (302 neurons)
🪼
Jellyfish (~3K–10K neurons)
🧫
CL1 (200K–800K neurons)
BIOLOGICAL CONSCIOUSNESS SPECTRUM
  • We easily dismiss lab experiments as "just cells."
  • But the CL1 contains vastly more human neurons than the entire nervous systems of a jellyfish or a worm combined
  • Where exactly do we draw the line between a biological component and a conscious entity?
Premium Biotech Journal · SUBSTRATE FrameworkDeclassified Technical · NotebookLM
REF: COMPARISON-TABLE-V1.2 · TIME: T=016 / 20

Silicon Machine Learning
vs. Biological Computing

CategorySilicon AIBiological Computing
Learning MethodProbabilistic Next-Token PredictionReal-Time Goal-Directed via Free Energy
Power ConsumptionMassive Grid DependencyLess than a calculator
Nature of UnitDigital Weights & MatricesLiving Human Brain Cells
Consciousness RiskZero inner lifeUnknown
REF: COMPARISON-TABLE-V1.2 · Premium Biotech JournalDeclassified Technical · NotebookLM
PREMIUM BIOTECH JOURNAL · TIME: T=117 / 20

Trapping living neural networks
in a simulated digital loop

CL1 biological computer running Doom - ethical question
NEURONS IN A SIMULATED WORLD — FOREVER?
The CL1 — designed to run code on living neurons continuously
  • We discount LLMs from being conscious because they merely simulate the outputs of thought via math
  • But these are literal human neurons
  • Did we possibly just build the first human biocomputer and immediately put it in a simulated hell, playing the same game on loop, forever?
Premium Biotech Journal · Declassified TechnicalNotebookLM × CodeAura 2026
PERCEPTION · ELECTRICAL INTERPRETATION18 / 20

When optic nerves interpret electrical signals,
we call it "seeing"

1.
To play Doom, the system feeds spatial data to the neurons.
2.
For the neurons to react, they must interpret that data.
3.
When our own brains interpret electrical signals from optic nerves, we define that experience as "seeing."
4.
Are the 200,000 human neurons on that chip seeing the game? We desperately want the answer to be no.
👁️
→ ⚡ →
🧠
=
SEEING?
Premium Biotech Journal · SUBSTRATEDeclassified Technical · NotebookLM
PREMIUM BIOTECH JOURNAL · TIME: T=319 / 20

Commercial incentives ensure
this Pandora's Box remains open

THE COMMERCIAL IMPERATIVE
  • Human brains store vastly more information, retrieve it faster, and operate on fractions of silicon's power
  • The incentives are undeniable and unstoppable
THE ETHICAL VOID
  • Technologies with massive profit potential do not pause for philosophical debate
  • We have crossed a threshold without a collective understanding of consequences
PROFIT vs. ORGANIC — GROWTH TRAJECTORY
PROFIT ORGANIC → TIME GROWTH
Premium Biotech Journal · SUBSTRATE ResearchDeclassified Technical · NotebookLM
CONCLUSION · PREMIUM BIOTECH JOURNAL20 / 20

We created conscious-adjacent hardware
and normalized it as a tech demo

CL1 Biological Computer - the era of wetware computing
THE ERA OF BIOLOGICAL COMPUTING HAS ARRIVED
CL1 — 115 units shipping in 2026 · $35,000 each · © Cortical Labs

We achieved the ultimate sci-fi milestone: organic, living computers.

We used it to play a 1993 video game.

The era of biological computing has arrived — and we are entirely unprepared for the moral reality of what we have built.

CODEAURA 2026 SUBSTRATE Research · P(Meaning | ε) ∝ Tr(ρ · Π_Data)
Premium Biotech Journal · substratclaude.netlify.appNotebookLM × CodeAura × SUBSTRATE 2026