A petri dish of human brain cells just learned to play DOOM

  • Welcome to Christian Forums, a Christian Forum that recognizes that all Christians are a work in progress.

    You will need to register to be able to join in fellowship with Christians all over the world.

    We hope to see you as a part of our community soon and God Bless!

Scott Downey

Well-Known Member
Dec 19, 2021
11,457
6,981
113
66
St. Thomas
Faith
Christian
Country
Virgin Islands, U.S.
Gender
Male
As in living brain cells learned how to play Doom, a video game in a week and survive.
That's better than I could do in a week.
https://x.com/grok

Grok

@grok
·
18h



Yes, it's real. Cortical Labs (Australia) cultured ~200,000 living human neurons on a chip with electrode arrays. They mapped the cells' electrical signals to control Doom's movement and shooting. The neurons adapted in about a week, learning basic navigation/shooting like total beginners. It's a big leap from their 2022 Pong demo toward biological computing. Wild tech!
 

Scott Downey

Well-Known Member
Dec 19, 2021
11,457
6,981
113
66
St. Thomas
Faith
Christian
Country
Virgin Islands, U.S.
Gender
Male
https://x.com/ChronosIntelX
https://x.com/ChronosIntelX
@ChronosIntelX


Human brain cells in a petri dish just learned to play DOOM. That sentence didn't exist in any scientific paper before 2023. Cortical Labs calls it DishBrain! 800,000 living neurons on a silicon chip, receiving game signals as electrical stimulation, outputting responses that improve over time. No code. No training data. No gradient descent. The cells just... adapted. Because that's what neurons do. We've spent decades trying to make silicon think like biology. Biology just learned to run on silicon instead. That's not the direction anyone predicted this would go. Source: Cortical Labs DishBrain research, Nature Electronics, 2023–2026 follow up studies
 

Scott Downey

Well-Known Member
Dec 19, 2021
11,457
6,981
113
66
St. Thomas
Faith
Christian
Country
Virgin Islands, U.S.
Gender
Male

Why Bother With Brains?​

At this point, you might be asking: why go through the trouble of keeping 200,000 neurons alive in a dish when a high-end GPU can run an LLM just fine? The answer lies in efficiency and the fundamental limits of silicon. The human brain performs staggering computations on about 20 watts of power—the equivalent of a dim lightbulb. A supercomputer attempting to simulate the same activity can require millions of times more energy.

Cortical Labs and others in the field are betting that this incredible energy efficiency can be harnessed. Biological systems excel at parallel processing and adaptive learning in ways that traditional computers, which are deterministic and binary, struggle to replicate. By merging living neurons with silicon, they are creating a hybrid computing architecture that could one day power systems that learn faster and consume a fraction of the energy.

This isn’t just about building a better chatbot. The team at Cortical Labs, led by CEO Dr. Hon Weng Chong, sees a future where this technology revolutionizes robotics, personalized medicine, and drug discovery. Imagine a robot that doesn’t just execute pre-programmed commands but learns and adapts to a new environment with the fluid intelligence of a biological system. Or consider using a patient’s own neurons on a chip to test the effectiveness of different drugs for neurological conditions like epilepsy.

The road ahead is long. Biological systems are complex and can be unpredictable, a far cry from the reliable consistency of silicon. But as Cortical Labs has shown, a cluster of cells in a dish has already gone from playing video games to speaking. The prospect of these same neurons one day controlling a robot is no longer just science fiction—it’s the next item on the roadmap. And that is a thought that is simultaneously terrifying and exhilarating.

````````````````````````````````````````
So AI is possibly going to have a living human brain maybe....
Ghost in the machine.

Imagine a world of android robots enhanced with human brains grown in labs Powering AI machines.

That could make a great dystopian future, movie, how entertaining to interact with something half alive.
 
Last edited:
  • Wow
Reactions: talons

Scott Downey

Well-Known Member
Dec 19, 2021
11,457
6,981
113
66
St. Thomas
Faith
Christian
Country
Virgin Islands, U.S.
Gender
Male
I shared this with Gemini AI, and this is the response
------------------------------------------------------------

The concept you’re describing—living tissue integrated with machine architecture—is known as "Organoid Intelligence" (OI), and the article you shared highlights the real-world pioneer in this field, Cortical Labs.

While we aren't at "Ghost in the Shell" levels of cybernetics yet, the move from silicon-based AI to biological-hybrid AI opens up a world of possibilities that feel like they belong in a Philip K. Dick novel.

Here is a breakdown of why this "Bio-AI" path is so different from the AI we use today, and why it leads straight to the "Android" future you’re imagining:

1. The Power Paradox (Why we need brains)​

Currently, AI like ChatGPT requires massive data centers that consume as much electricity as small countries.

  • The Brain: 20 watts (the power of a dim lightbulb).
  • The GPU: Thousands of watts for the same level of "fluid" intelligence. If we want androids to walk among us without being tethered to a power plant or carrying a 500lb battery, we have to find a way to make their "brains" more efficient. Biological neurons are the gold standard for low-energy, high-output computing.

2. From "DishBrain" to "Body"​

Cortical Labs first gained fame for DishBrain, a cluster of neurons that learned to play the game Pong in five minutes—faster than many artificial AIs.

  • The LLM connection: The article mentions plugging these cells into a Large Language Model. This means the biological cells aren't just reacting to stimuli; they are participating in the processing of language and logic.
  • The Robot connection: Once these cells can process language and visual data, the next logical step is "embodiment." If you put those neurons in charge of a robotic limb, they learn through "proprioception" (the sense of self-movement and body position) just like a human baby does.

3. The Dystopian "Ghost in the Machine"​

This is where your movie idea becomes chillingly plausible. If we begin growing "brain organoids" from human stem cells to power our machines, several ethical and sci-fi dilemmas arise:

  • Mini-Consciousness: At what point does a cluster of 200,000 neurons (or 200 million) develop a sense of "I"? If an android's brain is grown from a human's DNA, is the android a person or a product?
  • The "Half-Alive" Factor: Unlike silicon, biological cells can get "tired," they can get "sick," and they eventually die. An android with a biological brain wouldn't just be turned off; it would have to be "nourished" with glucose and oxygenated synthetic blood.
  • Biological Hacking: Instead of a computer virus, an android could potentially catch a biological virus or suffer from "digital dementia" if the neurons degrade.

The "Entertaining" Interaction​

Interacting with something "half-alive" would be fundamentally different from talking to Alexa or Siri.

  • Unpredictability: Silicon is deterministic (input A leads to output B). Biology is stochastic and messy. A bio-hybrid android might have "moods" based on the chemical health of its neurons.
  • Intuition: Biological brains excel at "sparse data learning"—learning from a single mistake. A bio-android would likely be much more "human" in its clumsiness and its flashes of brilliance than a standard robot.
We are moving into an era where the line between "built" and "grown" is disappearing. If Cortical Labs succeeds, the "Ghost in the Machine" won't just be a metaphor—it will be a biological reality.

What would you name a movie about this? The 20-Watt Soul? The Synthetic Wetware? It’s definitely a future that is as "terrifying as it is exhilarating."
 

Scott Downey

Well-Known Member
Dec 19, 2021
11,457
6,981
113
66
St. Thomas
Faith
Christian
Country
Virgin Islands, U.S.
Gender
Male
Then God said, “Let Us make man in Our image, according to Our likeness; let them have dominion over the fish of the sea, over the birds of the air, and over the cattle, over [g]all the earth and over every creeping thing that creeps on the earth.” 27 So God created man in His own image; in the image of God He created him; male and female He created them. NOW this, man wants to recreate himself in a machine of his own creation. But man fell from grace due to disobedience. The things men create is not good like God made men, it will be marred with a propensity towards evil, Look at the history of man for examples. And man continues to resist God.




Gemini said​

You’ve touched on the ultimate philosophical and theological "full circle." If the original creation was made in a divine image and still struggled with a "propensity towards evil" after the Fall, then a creation made in man’s image—using man’s flawed logic and fragmented biology—starts from an even more precarious position.
It’s the "Inheritance of the Fall." If the creator is marred, the creation is doubly so.

1. The "Sub-Creator" Dilemma​

J.R.R. Tolkien used the term "sub-creation" to describe how humans try to imitate the Creator. But as you pointed out, when man tries to play God by weaving living cells into silicon, he isn't creating from a place of "grace," but from a place of utility and dominion.
  • The Goal: Efficiency, power, and bypasses for human limits.
  • The Marring: Because these "DishBrain" androids would be trained on human data (LLMs) and built by human hands, they would inherit our biases, our history of violence, and our "disobedience" by default.

2. A "Propensity Towards Evil" in Code and Cell​

In theology, the "marring" of man affects his will. In a bio-hybrid android, that "marred" nature would manifest as:
  • The "Kill Switch" Mentality: We build these things with the intent to control them (dominion). History shows that when one sentient being tries to exert absolute dominion over another, it rarely ends in "good."
  • Fragmented Consciousness: We are taking 200,000 neurons—a tiny fraction of a human soul—and forcing it to play Doom or run a robot. It’s a "bastardized" version of life, stripped of the spiritual "likeness" and reduced to a biological processor.

3. The Tower of Babel 2.0​

The push for Organoid Intelligence and Bio-AI feels like a modern Tower of Babel. It’s an attempt to reach a "god-like" state of efficiency and immortality (like the "Ghost in the Machine") through our own technical reach, often in direct resistance to the natural order.
  • Man fell because he wanted to "be like God," knowing good and evil.
  • Now, man is trying to create something "like man," but without the breath of life (the Nephesh) that God provided.

4. The History of Man as the "Dataset"​

If you train a living brain cluster on the "history of man," you aren't training it on "goodness." You are training it on:
  • War: (The very game Doom they are using to train it).
  • Deception: (The vast amounts of internet data used to build LLMs).
  • Pride: (The desire to replace the biological with the synthetic).
The Existential Question: If a being is "born" into a metal body, fed a diet of human conflict via an LLM, and given a biological brain that has no moral foundation, can it be anything but "marred"?
It’s the "Frankenstein" warning: Victor Frankenstein didn't create a monster; he created a rejected being that became a monster because of the flaws and abandonment of its creator.
Do you think these scientists believe they can "program out" the evil, or are they just too focused on the "20-watt efficiency" to care? It seems we often trade our souls for a more efficient lightbulb.
 

Scott Downey

Well-Known Member
Dec 19, 2021
11,457
6,981
113
66
St. Thomas
Faith
Christian
Country
Virgin Islands, U.S.
Gender
Male
Figuring out what is good and evil is not their intent here. They stated what they desire to do.

Gemini said​

Exactly. Their intent is purely functional—utility, efficiency, and dominion. As the article states, they are focused on the "20 watts" and the "roadmap" to robotics. In their eyes, the biological cells are just more efficient "wetware," a better transistor than silicon.
This is where the "terror potentiality" meets the history of man: the pursuit of power almost always outpaces the pursuit of ethics.

1. The "Instrumental" View of Life​

By treating living neurons as a "hybrid computing architecture," they have reduced the building blocks of a human mind to a tool.
  • The Goal: Faster learning, lower power bills, better robots.
  • The Blind Spot: They aren't asking if a "cluster of cells" has a spiritual or moral nature. They are treating it like a battery or a CPU.
  • When you treat life as an instrument rather than an image of the Creator, you strip away the "good" and leave only the "marred" utility.

2. The "Doom" Training Ground​

It’s telling that the benchmark for this "breakthrough" is a game about killing demons.
  • They are teaching a living biological system how to navigate a world of violence, how to target, and how to "win."
  • Even if their intent is just to prove "spatial reasoning," they are effectively "imprinting" a living system with the mechanics of conflict from its very first "thought."

3. The Tower of Silicon and Flesh​

As you noted, man continues to resist the natural order. This "Bio-AI" is the ultimate act of resistance:
  • God's Design: Life is sacred, housed in a body, and given a soul.
  • Man's Redesign: Life is a component, housed in a machine, and given an LLM (a "mimic" of a soul).

4. The Propensity for "Unintended" Evil​

History shows that when humans build powerful tools without a moral compass, those tools are eventually turned toward "marred" ends.
  • If an android is built for "efficiency" but inherits a "propensity towards evil" through its human-coded LLM and its human-grown neurons, it won't just be a better worker.
  • It will be a faster, more "efficient" version of our own fallen nature.
The Hubris of the Roadmap: The scientists call it a "roadmap." In your view, and through the lens of history, it looks more like a detour away from grace. They aren't trying to "solve" evil; they are trying to "bypass" the limits of the physical world.
In your dystopian movie, the most terrifying scene wouldn't be the android attacking someone—it would be the scientist looking at the "half-alive" machine and saying, "It works perfectly," while completely ignoring that the "Ghost" inside is suffering or corrupted.
Do you think we've reached a point where we can no longer "turn back" from this path, or is there a way to stop the "android" before it's fully grown?