Yes, the demonstration is real—but “human neurons learned to play Doom” is a misleading shorthand. Cortical Labs connected living human-derived neurons to its CL1 biological-computing platform and a computer running Doom. Electrical stimulation represented parts of the game state; recorded neural activity was decoded into actions such as movement, turning, and shooting.
The result was a striking feasibility demonstration of closed-loop biological computing, not an intact human brain mastering a first-person shooter. The character wandered, fired ineffectively, and died frequently. The neurons did not see a monitor, understand demons or weapons, or operate a complete computer by themselves.
| # | Preview | Product | Price | |
|---|---|---|---|---|
| 1 |
|
DOOM Eternal: Standard Edition - PlayStation 4 | $27.49 | Buy on Amazon |
| 2 |
|
DOOM: The Dark Ages – Xbox Series X | $39.99 | Buy on Amazon |
| 3 |
|
DOOM: The Dark Ages – PlayStation 5 | Buy on Amazon | |
| 4 |
|
Doom - Xbox One | $26.99 | Buy on Amazon |
| 5 |
|
DOOM + DOOM II (Limited Run Games #144) - for Playstation 5 | $44.48 | Buy on Amazon |
As an Amazon Associate I earn from qualifying purchases.
What actually happened?
Cortical Labs used a culture of approximately 200,000 living human-derived neurons in a system connected to a digital copy of Doom, according to Scientific American’s reporting. The game still ran conventionally in software. The biological culture formed one part of the input, processing, and control loop.
Recommended Free Tools
Software extracted relevant information from the game, such as the direction of an enemy or the location of a target. That information was converted into electrical stimulation patterns delivered through electrodes. The neurons responded with electrical spikes, which were recorded and mapped back into game controls. Those controls changed the game state, producing new feedback for the culture.
#1 Best Overall
- Gain access to the latest demon-killing Tech with the DOOM Slayer's advanced praetor suit, including a shoulder-mounted flamethrower and the retractable wrist-mounted DOOM Blade
- Upgraded guns and mods, such as the Super shotgun's new distance-closing meat hook attachment, and abilities like the double Dash make you faster, stronger, and more versatile than ever
- You can't Kill demons when you're Dead, and you can't stay alive without resources. These tools are the key to your survival and becoming the ultimate demon-slayer
- A new class of (destructible) demon
- Battle mode is the new 2 versus 1 multiplayer experience built from the ground up at id software
The loop can be summarized as:
Doom state → electrical stimulation → neural activity → spike decoding → game action → new feedback
This is why the most accurate description is that the neurons participated in controlling Doom. They did not render the game, run the operating system, or independently decide how to play it.
Did the neurons learn to shoot?
In a limited operational sense, yes. The culture adapted its activity patterns in a way that could produce control signals associated with actions including shooting. But that does not show that the neurons understood guns, enemies, goals, or the fictional world of Doom.
Free tools Windows power users keep installed
One-click scans. No signup required.
“Learned to produce shooting-related control signals” is therefore safer than “understood how to shoot.” The reported gameplay was clumsy: the character moved through the environment, encountered enemies, fired, and died often. It was nowhere near human play or the performance of modern game-playing software. Independent coverage describes the demonstration’s frequent failures and limited behavior.
How feedback shaped the behavior
The public explanation of the experiment describes a feedback-driven system. Structured, predictable feedback represented successful behavior, while unpredictable electrical noise indicated poor or unsuccessful behavior. Over time, the culture’s activity changed in response to that signal.
That is broadly comparable to reinforcement-like adaptation, but it is not the same as giving a person a written lesson or showing an artificial intelligence a labeled training set. The network was embedded in a carefully designed interface that determined what information it received, how success was represented, and how its activity affected the game.
The biological contribution is meaningful precisely because it was inside a real-time closed loop. But the loop was engineered by humans and supported by conventional software and electronics.
Were the cells taken from someone’s brain?
No. The neurons were human-derived, but they were not removed from an intact human brain. The reported process starts with cells from adult donor skin or blood. Those cells are reprogrammed into induced pluripotent stem cells and then differentiated into neurons. Scientific American outlines this origin.
Rank #2
- Developed by id Software, DOOM: The Dark Ages is the prequel to the critically acclaimed DOOM (2016) and DOOM Eternal that tells the epic cinematic origin story of the DOOM Slayer’s rage.
- In this third installment of the modern DOOM series, players will step into the blood-stained boots of the DOOM Slayer, in this never-before-seen dark and sinister medieval war against Hell.
- A dark fantasy/sci-fi single-player experience that delivers the searing combat and over-the-top visuals of the incomparable DOOM franchise, powered by the latest idTech engine. With a customizable difficulty system, it’s the perfect entry point whether you’re new to the franchise or a long time fan.
- As the super weapon of gods and kings, shred enemies with devastating favorites like the Super Shotgun while also wielding a variety of new bone-chewing weapons, including the versatile Shield Saw.
- Experience the origin story of the DOOM Slayer’s rage in this epic, cinematic, and action-packed story.
Calling the setup “a human brain in a dish” is vivid but inaccurate. A neural culture does not have a complete brain’s anatomy, sensory organs, body, developmental history, or normal connections to the outside world. It is a living network of neurons grown under laboratory conditions and connected to electrodes.
What is the CL1?
Cortical Labs describes the CL1 as a code-deployable biological computer. Living neurons grow on a silicon microelectrode array. The platform stimulates the cells, records their electrical activity, and includes integrated life-support hardware for maintaining the culture.
The company says the system can keep neurons alive for up to six months and supports bidirectional communication: electrical signals go into the culture, and neural signals come back out. IEEE Spectrum reports that the CL1 uses a 59-electrode interface, compared with eight inputs in the earlier DishBrain prototype, and operates with sub-millisecond feedback loops. Those specifications should be treated as company and product reporting, not as independently audited benchmarks. IEEE Spectrum’s engineering overview provides the relevant context.
The Tool Desk
Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →The approximately 200,000 neurons reported for the Doom demonstration should not be confused with the CL1’s advertised capacity. IEEE Spectrum has described the commercial platform as supporting approximately 800,000 neurons. One number refers to the reported demonstration culture; the other refers to the stated platform capacity.
Although the CL1 is described as self-contained for its core biological operation, experiments still need conventional computers and software to create stimuli, run interfaces, decode spikes, and present tasks. It is not a general-purpose PC made entirely from neurons.
From Pong to Doom
The Doom demonstration follows Cortical Labs’ earlier DishBrain work. In a peer-reviewed 2022 paper in Neuron, researchers connected cultured neurons to a simplified Pong environment and reported learning-like adaptation within minutes. The published DishBrain study is the strongest scientific foundation for the earlier experiment.
Pong is a highly constrained two-dimensional task involving a paddle and a ball. Doom adds navigation, orientation, enemies, targets, movement through a richer environment, and shooting. That makes it a more memorable demonstration of a biological network operating in a complex feedback loop.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
It does not, however, prove a straight progression from Pong to general intelligence. Both tasks remain heavily engineered, and the neurons operate inside a substantial digital and hardware stack. The significance is that the same broad idea—cultured neurons receiving structured input and influencing an environment—can be extended to a more demanding task.
Rank #3
- Developed by id Software, DOOM: The Dark Ages is the prequel to the critically acclaimed DOOM (2016) and DOOM Eternal that tells the epic cinematic origin story of the DOOM Slayer’s rage.
- In this third installment of the modern DOOM series, players will step into the blood-stained boots of the DOOM Slayer, in this never-before-seen dark and sinister medieval war against Hell.
- A dark fantasy/sci-fi single-player experience that delivers the searing combat and over-the-top visuals of the incomparable DOOM franchise, powered by the latest idTech engine. With a customizable difficulty system, it’s the perfect entry point whether you’re new to the franchise or a long time fan.
- As the super weapon of gods and kings, shred enemies with devastating favorites like the Super Shotgun while also wielding a variety of new bone-chewing weapons, including the versatile Shield Saw.
- Experience the origin story of the DOOM Slayer’s rage in this epic, cinematic, and action-packed story.
Is this artificial intelligence?
The best terms are biological computing, wetware computing, or hybrid biological-silicon computing.
- The game engine is digital.
- Software encodes game state into stimulation patterns.
- Specialized electronics stimulate and record the culture.
- Software decodes neural spikes into actions.
- Conventional computing remains part of the complete system.
A useful formulation is: the neurons are an adaptive biological processing element inside a hybrid computer. Calling the entire arrangement “AI” is not necessarily wrong in the broadest sense, but it hides the important division of labor between living tissue, silicon hardware, and software.
Does this prove consciousness or sentience?
No. Nothing in the Doom demonstration establishes subjective experience, consciousness, suffering, or a desire to play.
Cortical Labs’ earlier company-linked paper used “sentience” in its title, but that wording is controversial and should not be treated as a settled scientific conclusion. Adaptive behavior can be described at several levels:
- Learning-like adaptation: a change in activity associated with feedback.
- Goal-directed behavior: a description of system-level behavior, not proof of conscious goals.
- Agency: a debated theoretical interpretation.
- Consciousness or sentience: not established by this demonstration.
The culture is not an animal, a person, or an insect. Anthropomorphic language makes the headline more exciting, but it makes the science less precise.
Why use neurons instead of conventional silicon?
Researchers are interested in neural cultures because biological networks may offer useful properties that are difficult to reproduce directly in conventional hardware. Potential areas of investigation include:
- Adapting from relatively little training data.
- Responding to changing inputs.
- Studying human neural processing directly.
- Testing how drugs alter network behavior.
- Modeling neurological disease using different donor-derived cell lines.
- Building adaptive controllers for robots or other embodied systems.
- Exploring biological computation as a complement to digital architectures.
Cortical Labs’ research index lists work involving neural cultures, drug treatment, sample efficiency, disease-related function, and biological intelligence.
Energy efficiency is often presented as a major attraction. The human brain’s roughly 20-watt energy use is a popular comparison, but it is not an apples-to-apples benchmark against a purpose-built silicon workload. A fair accounting would include the culture’s life support, stimulation and recording electronics, control systems, external software, and any laboratory infrastructure. Energy claims should therefore be treated as comparisons or company estimates unless a standardized system-level measurement is provided.
Rank #4
- A Relentless Campaign: There is no taking cover or stopping to regenerate health as you beat back Hell's raging demon hordes
- Return of id Multiplayer: Dominate your opponents in DOOM's signature, fast-paced arena-style combat
- Near-Limitless Gameplay: Doom SnapMap – A Powerful, but Easy-to-Use Game and Level Editor That Allows for Limitless Gameplay Experiences on Every Platform
- Entertainment Software Rating Board (ESRB) Content Description: Blood and gore, intense violence, strong language
What could the platform be used for?
The strongest near-term case is research rather than entertainment. A biological-computing platform could help laboratories:
- Study neural cultures while they process information in an embodied environment.
- Compare networks derived from different donors.
- Investigate epilepsy and other neurological disorders.
- Measure how candidate drugs affect learning and network dynamics.
- Explore how living networks adapt under controlled stimulation.
- Test hybrid biological-digital controllers in robotics.
The commercial CL1 is aimed at universities, biotechnology companies, and neuroscience laboratories—not ordinary gamers. Cortical Labs also offers Cortical Cloud, which is intended to provide remote access to biological-computing infrastructure without requiring a customer to maintain living cultures.
For technical researchers, the company publishes an early CL API documentation repository with Python-oriented examples and notebooks. The repository describes the material as an early or pre-release look, with important documentation still incomplete.
Do these 3 things before closing this tab:
1Scan for outdated or missing drivers - takes under a minute2Repair Windows errors before they cause bigger problems3Fix the driver behind crashes, sound loss and screen glitchesCould you buy one?
Cortical Labs has a CL1 purchase page, but the reviewed official page did not show a plainly visible current public price. A figure of approximately $35,000 has appeared in secondary coverage, including IEEE Spectrum, but it should be treated as a dated reported price rather than a confirmed current quote.
More importantly, buying the hardware is not like buying a desktop PC. Physical-unit users need suitable cell-culture facilities, biosafety procedures, technical expertise, and appropriate ethical oversight. IEEE Spectrum reports that the system is not intended for casual home operation.
The practical options are:
| Option | Advantage | Trade-off |
|---|---|---|
| Buy a CL1 | Maximum control over experiments | Capital cost, laboratory infrastructure, cell maintenance, and oversight |
| Use Cortical Cloud | No need to maintain a physical culture | Access, availability, and pricing may require signup; it is not a gaming or GPU-cloud replacement |
| Read the API documentation | Low-commitment technical evaluation | Documentation is early and not necessarily production-ready |
Ordinary gaming PCs, GPUs, consumer brain-computer interfaces, and neuromorphic development boards may be useful comparison technologies, but they are not direct substitutes for a living-neuron platform.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.What the demonstration does not show
- It is not a standardized benchmark. The public demonstration does not provide the kind of methods, controls, metrics, and reproducibility data expected from a full benchmark paper.
- It is not a miniature human brain. Neuron count says little by itself about organization, connectivity, or capability.
- It is not autonomous in the ordinary sense. The culture depends on designed stimulation, decoding, software, electronics, and a digital environment.
- It does not beat modern game-playing AI. The visible performance was poor compared with even basic software agents.
- It does not establish general intelligence. A culture adapting to a narrow task cannot support claims of human-level intelligence or imminent AI replacement.
- It does not prove low total energy use. Life support and supporting infrastructure must be included in any serious comparison.
- It does not remove biological variability. Cultures differ between preparations, require calibration, and have finite viability.
As of August 18, 2026, the strongest evidence for the Doom result is Cortical Labs’ demonstration and explanatory material, supported by independent reporting. Cortical Labs’ research page lists peer-reviewed work on related systems, including DishBrain, but no peer-reviewed paper specifically documenting the Doom experiment was identified in the reviewed sources. It should therefore be described as a company demonstration, not yet as a peer-reviewed Doom-specific breakthrough.
Outdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware matchPC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11The ethical questions are becoming more practical
Even without evidence of consciousness, increasingly complex neural cultures raise legitimate governance questions. Researchers and regulators may need to consider:
Best Value
- DOOM + DOOM II on a region-free physical disc.
- Includes: DOOM, DOOM II, TNT: Evilution, The Plutonia Experiment, Master Levels for DOOM II, No Rest for the Living, Sigil & Sigil II, Legacy of Rust (a new episode created in collaboration by id Software, Nightdive Studios and MachineGames).
- A new Deathmatch map pack featuring 25 maps
- Total of 187 mission maps and 43 deathmatch maps in DOOM + DOOM II
- # of Players: Single System 1-4, Local wireless 1-8, Online 1-16
- What standards should apply to experiments involving human-derived neural tissue?
- What consent should donors provide for reprogrammed cells and future uses?
- At what level of complexity might a culture acquire morally relevant properties?
- How should cultures connected to robots or other embodied systems be classified?
- Does describing a system as “animal-free” mistakenly suggest that all ethical concerns have disappeared?
These are open questions, not evidence that the CL1 is conscious or suffers. The sensible approach is to improve oversight as the systems become more complex, rather than waiting for a definitive philosophical answer.
How to judge claims about biological computers
When a future headline says that neurons learned a new task, ask:
- What did the neurons actually do? Separate biological processing from digital preprocessing and control.
- Was the loop genuinely closed? The culture should receive meaningful feedback and influence the environment in real time.
- Did performance improve? Look for trial-by-trial results and a clear baseline.
- Can others reproduce it? Biological variability makes independent replication especially important.
- How difficult was the task? Identify what the environment simplified or preprocessed.
- What energy was counted? Include life support, interfaces, electronics, and external compute.
- Is there a full publication? Methods, data, controls, and statistics matter more than a video.
- What practical advantage was demonstrated? Feasibility is different from outperforming existing systems.
Frequently Asked Questions
Did a human brain play Doom?
No. A culture of human-derived neurons was connected to a digital Doom environment through electrodes, software, and recording hardware. It was not an intact brain and did not independently operate a computer.
Quick wins for a faster PC:
Scan for outdated or missing drivers - takes under a minuteDriver Scan →Repair Windows errors before they cause bigger problemsFix Now →Were the neurons conscious?
The demonstration provides no evidence of consciousness, subjective experience, suffering, or a desire to play. It supports learning-like adaptation in a designed feedback loop.
Is the CL1 a replacement for a gaming PC or GPU?
No. The CL1 is a specialist biological-computing research platform requiring living-cell maintenance and substantial experimental infrastructure.
Can researchers access the system without buying the hardware?
Cortical Labs presents Cortical Cloud as a remote-access option. Current availability, pricing, and terms should be checked through the official signup flow.
The Bottom Line
Cortical Labs’ Doom demonstration matters because it shows living human-derived neurons influencing a more complex digital environment through a real-time feedback loop. It does not show a brain in a dish, conscious gameplay, general intelligence, or a biological replacement for silicon computers. For now, the credible opportunity is as a research platform for neural processing, disease, drugs, and adaptive hybrid computation.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
