Brain Organoids, mage credit, NAID, https://www.flickr.com/photos/niaid/

Discover the world of lab-grown human ‘mini-brains’ and if whether they could lead towards a new sci-fi apocalypse or medical marvels.

The human brain is nature’s greatest machine. A supercomputer floating around in our skulls making us capable of sentience, art, imagination; and everything that makes us human, all whilst consuming less power than your toaster. And what is AI but mankind’s shoddy facsimile of the natural symphony of firing neurons, absent of the unknown complexities that give us life? But part of nature is its imperfections and even the marvel of the mind is still vulnerable to disease, fate and time.

Around the world, scientists are crafting ‘miniature human brains’ in petri dishes. Some even integrate them with AI to create what sounds like something every piece of science fiction media warns against creating.  It’s as if all those cautionary tales flew straight over their heads while they put them in computer simulations like in The Matrix or create chimaeras by injecting them into the skulls of mice but will we end up with a rebellious lab-grown super intelligence, the planet of the rodents, or will we finally be able to treat once incurable diseases or process data like never before. Luckily for us, the truth is more exciting than frightening.

So what exactly are these mini-brains? Creating Brain organoids starts with a dish containing hundreds of tiny compartments, each holding stem cells that are stimulated to then grow into brain cells. Due to not being connected to a human body, they lack the circulatory systems to stay alive very long or get very big. When they say ‘mini’, they mean it, as most only grow to 1 or 2mm. The challenging part is maintaining them. While making thousands of organoids can take just 30 minutes, they need up to 200 days to mature to the point where they resemble early brain development.

But what is the actual point of these cranial cocktails?

One avenue is utilising the brains for computing and processing data with AI and despite their limitations, scientists have done incredible things. Cortical labs in Australia even used chemical and electrical inputs to train the organoids to play pong in a computer simulation like some sort of proto-matrix.

Dr Fred Jordan the co-CEO of Final Spark a Swedish company that integrates organoids into computer chips that researchers can use to experiment with and improve machine learning, says organoids will revolutionise generative AI.

“Artificial neural networks…AI works by simulating neurones. And, therefore, it’s quite an obvious statement that, if I replace a simulation with the real thing, it is going to work a little bit.” He said.

Dr Thomas Hartung has five different professorships at three universities including Johns Hopkins University. He is part of the Centre for Alternatives to Animal Testing and is interested in exploring organoids and their ethical implications.

Currently, these neurons are sitting bored, kept in cell cultures eager to find something to think about.

Dr Hartung says: “They are kept as happy as we can but they have no idea about what is happening around them but these neurons they start talking to each other and the basic idea of organoid intelligence was if we give them input what happens if we start communicating”

Scientists are working on larger organoids with blood support, but for now, the focus is on small ones. This size limit is crucial as it prevents any concerns about self-consciousness or other advanced functions.

But these organoids are pretty useless without training—it’s all about sending the right signals and tossing them a “treat” when they cooperate, like a weird dog. Scientists reward the mini-brain with chemicals like glutamate when it plays nice, really driving home the whole “living” computer idea.

Dr Paul Holloway, a researcher at the University of Oxford Studying stroke and Neurovascular disease explained a large organoid is typically about five or so millimetres cubed.  

“This is like 50 times larger than a fruit fly’s brain. But a fruit fly you wouldn’t really consider to be sentient.” He said.

In this cubic millimetre of human brain tissue, there are about 90,000 neurons and 700 million synapses. Although this sounds impressive, don’t forget that the modern-day M1 chip from Apple has 114 billion transistors.

In biology, information is processed freely within neurones, which are more complex than transistors, so they can’t be considered like for like. So although there is processing power, these organoids’ lack of natural inputs and outputs and their small size make it a bit of Science Fiction that they’re actually thinking.

Dr Jordan says “Compared to a modern computer it’s perfectly useless” but in his mind, it is just a matter of time and a necessary avenue considering the poor energetic efficiency of traditional AI. 

To simulate our brains it would take something like a small nuclear plant. 100 billion neurons each with 10000 connections, but right now it’s sitting in your skull consuming 20 watts.

“It’s a no-brainer. It’s the right path and of course, particularly, this is something that would shock a number of people. And I think this is probably one of the reasons this has not been done before. Because I think society just becoming ready to accept such an idea. Because the idea was around for 50 years, it’s not new but there is always a difference between people dreaming of things and people doing things” Dr Jordan said.

Alexia King is the supervisor for the Brain Organoid Hub at Emory University, a collective of

Scientists, who generate quite thousands of organoids per month and act as a resource for other Labs. For her and many others these organoids most practical uses are in medicine.

She says: “This is the closest that we have to doing research on humans without actually having to use a human.”

These human cell-based systems allow scientists to study human cells in a more relevant context. The brain is more than neurons, Brains contain all sorts of cells and 400 miles of blood vessels.  Organoids contain these things meaning scientists can see the interaction between these cells and it can allow them to study diseases, brain function and new drugs, and most importantly on human brain cells, relying less on animal models, because for obvious reasons it’s a bit difficult to get access to functioning human brain tissue ethically unless you want to become the subject of another true-crime podcast.

These scientists understand this technology can inspire visions of nightmarish futuristic worlds of sentient lab-grown brain-powered supercomputers. Luckily unlike those in Jurassic Park, most scientists do think about consequences.

Dr Hartung says: “We are very far away from anything a handheld calculator could do…but it’s always a beginning. If you have technologies that are growing exponentially and you combine several of these technologies, it could be very fast that you see miracles happening. And this is why we are trying to have this discussion early on here.

Each and every scientist is different, for me ethics are fundamental in research… Because I mean you have to imagine I’m a physician. I want to take care of people. But at the same time, I don’t want to disrespect the animals “

Researchers at Stanford have even implanted small brain organoids into rats, and they observed that the organoids integrated into the animals’ brains. This is possible because the brain lacks immune cells, so there’s no rejection. While this raises ethical concerns, it also opens up possibilities for regenerative medicine. For example, a patient’s brain tissue could potentially be used to repair damage from a stroke and restore lost functions.

“So this research makes sense. It is not just to create some homunculus…But you have also to ask how far you want to go. Because when you can move from the rat to the monkey, you can make this brain organoid much bigger very quickly and then you suddenly have a monkey human and you don’t know what happens.” Hartung says.

But whose responsibility is it to ensure we don’t accidentally create a sci-fi apocalypse while playing with our intelligent goo?

Hartung says: “It’s not that easy because many people tend to see a technical problem and want to solve it and if they’re not asked particularly to think about the ethical problems and start the discussion with ethicists… They can sometimes forget about these aspects and just look into achieving their technological goal. So to some extent for the bigger questions we need also society to step in and science organisations.”

Maybe it’s just human nature to look for danger, to jump to conclusions. Our brains can conjure up all sorts of stories but maybe it’s still important to keep one foot in reality.

Alexia King says: “Even I am trying to kind of step away from saying mini-brains or things like that because I think it does paint a picture that we are creating some Frankenstein kind of thing in the lab… I hope that we can at some point step away from the sensationalism of organoids and really talk about what’s actually going on in the dish.”