I've Got Questions with Sinead Bovell - Brain Scientist On AI: You Aren’t Ready For What’s About To Happen | Dr. Alysson Muotri

Episode Date: November 27, 2025

Dr. Alysson Muotri is a world-leading geneticist and pioneering neuroscientist whose work pushes the boundaries of how we understand the human brain. His lab grows brain organoids—living “mini-brains” created from stem cells—to study how the human mind develops, ages, breaks down, and how it might one day be repaired. In this episode, we get into why he’s trying to reconstruct the human brain in a lab, how brain organoids are transforming our ability to study and potentially reverse brain-based illnesses, and the emerging field of biocomputing and “organoid intelligence.” We explore how these living neural systems could shape the next era of AI, what it would mean for AI to be powered by biological tissue, and how this convergence of biology and computation could reshape our future. Follow my work here:Website: ⁠⁠⁠⁠⁠https://www.sineadbovell.com⁠⁠⁠⁠⁠ Substack: ⁠⁠⁠⁠⁠https://sineadbovell.substack.com⁠⁠⁠⁠⁠ Instagram: ⁠⁠⁠⁠⁠https://www.instagram.com/sineadbovell⁠⁠⁠⁠⁠ LinkedIn: ⁠⁠⁠⁠⁠https://www.linkedin.com/in/sineadbovell⁠⁠⁠⁠⁠ Twitter / X: ⁠⁠⁠⁠⁠https://twitter.com/SineadBovell⁠⁠⁠⁠⁠ YouTube: ⁠⁠⁠⁠⁠https://www.youtube.com/Sineadbovell⁠⁠⁠⁠⁠ TikTok: ⁠⁠⁠⁠⁠https://www.tiktok.com/@sineadbovell⁠⁠

I'm someone trying to reconstruct the human brain in the lab. But what does it actually mean to build a brain in a lab? If I take your skin cells, I can reprogram the cells into the stem cells. From there, it started recreating your brain. And then you could predict, for instance, what neurological diseases I may be at risk of. That's why I'm applying that to understanding, for example, autism or epilepsy. Can we interfere? Are they reversible? The brain is wired to become conscious.

And if we start using them to power AI. Is there a world in which your personal AI system is maybe also powered by your personal brain cells? Yeah, I think so. People sometimes say we can't disprove or prove that current AI systems are conscious. But an AI system powered on human brain cells. If we start using the algorithms coming from the organic cells, I think it's going to be inevitable. We are moving into a cyborg. Welcome back to I've Got Questions.

I'm your host, Sheneuveauvel, and you're joining me here in Sandy, for a conversation with Dr. Alison Motri. Dr. Motri has one of those jobs where you never knew it existed until you hear him describe it. And this is going to be one of those conversations where you think, wow, I see how it's going to change the world. Dr. Montri, how would you describe the research and the work that you do? That's a very provocative question. I'm someone trying to reconstruct the human brain in the lab because we don't have really good treatments for neurological conditions.

So if you ask, and I did that myself, and I invite you in your audience to do the same, if you ask like a neuropsychiatry or a neurologist or a neuropediatric, how many people did you cure? And most likely the answer is zero. And that's because of the inaccessibility of the human brain. brain that starts in neuterous. So we don't have good tools with the right resolution to actually understand how the brain is formed cell by cell, synapsy by synapsy, neuron by neurons. So there is no such

a tool. And then for the rest of your life, it remains inside of your skull. And again, I mean, the tools that we have doesn't have the right resolution to study. So we have no idea how the brains form. And that's why we don't understand how neurological conditions happens. And we don't know how to treat them. We can manage the symptoms, but no cure. And these in contrast, we, for example, the heart, where we can study the heart, your lungs, your blood, everything is accessible, but not the brain. And I guess when people here reconstructing the brain, how does one even do that? I mean, I'm sure people are picturing a scalpel in the brain and taking out a few cells, but what does it actually mean to build a brain in a lab?

That's a great question. So the role. The core material is what we call a pluripotent stem cell. So this is a kind of a stem cell that can differentiate or specialize. And what would be a stem cell? For anyone who has missed that grade 10 lecture, what is the stem cell? A stem cell is a type of cell that self divides, so it survives indefinitely because keeps dividing. But at the same time has the potential to specialize in a specific tissue, your skin,

your hair cells, your brain cells. So these stem cells, they have, they are what we call undifferentiated. They are not committed yet to any tissue. So we are learning, that's what my lab does, we are learning ways to drive these cells to commit to the neurolineage so they can start building the brain. So a stem cell is a cell that can become any cell. And so you could take, for example, a skin cell from somebody, turn that into a stem cell and then kind of coax

that stem cell along to becoming a brain cell? That's correct, yeah. And then how involved do you have to be to coax them along to start to behave like the brain cells? Are you very involved? Is there serums you're adding? Or do they just become brain cells on their own? Yeah, so there are two

steps. The first one is where you actively add factors into these tissue culture formulation, and these are the factors that will induce them to become brain cells and not a skin cell or lung cell, right? And so we do that at the same time that we use factors that avoid the specialization in

other tissues. So we are favoring the brains. So that requires you to actively add the specific factors at a certain time points. But once you pass this phase, then the cells do by themselves. So it's all genetically encoded. So they self-assemble in a three-dimensional tissue. And this self-assemble, it's, again, I mean, it's all written in your DNA. The cells know how to divide, where to migrate, which cells to make connections with the neighbors,

and it starts forming neural circuitries. And it starts everything a little bit randomly. Neurons at one point start firing, and they will make connections with other neurons nearby, forming microcircuitedries. At one point, that might take four or five months. these microcircuaries start to talk to each other inside this tissue. And after that, the complexity of the circuitries just increase. You start seeing different frequencies, different intensities of signaling coming.

And again, everything spontaneously, everything genetically encoded. So we don't have to do much. So the neurons are firing, synapses are forming. How many neurons? We have billions of neurons, right? So how big are these brain cells or how big is this structure growing in the left? So right now, the tissue that we make can grow up to 0.5 centimeter. So it contains about 2.5 million neurons in there.

Usually we have 5 million cells because there are other cells in the brain that are not neurons. So the tissue will grow up to 0.5 centimeter in diameter. So it's a small tissue compared to the brain. Orders of magnitude is smaller. And the number of neurons is also reduced to 0.5 million neurons versus billions of neurons in the human brain. And the reason why we cannot scale it up is because these tissues are not vascularized yet. We haven't figured out how to induce the vascularization as the brain is formed. Vascularization is getting blood flow and nutrients and oxygen to these cells.

Yeah, yeah. And because it's a two-dimensional tissue, all the surrounding cells have gotten. getting their nutrition by diffusion of the formula that they are in there. But the center of these spheres, it's a sphere form, we will end up becoming necrotic because the nutrients are not getting there. So we are still learning how to vascularize. Once we learn how to vascularize, then yes, then we can have even though like a blood flow and nutrients can get to the inside of the tissue and they will eventually grow bigger.

And so how long would current brain organoids last for? Could they live the lifespan of a human in the lab or how long could they be alive? Yeah, most likely. We have kept them just to answer those questions for several years. And we just stopped the experiments because someone dropped the plate. So it was an accident. But we were able to keep them for three to five years. And how functionally similar are the brain cells in the lab to the lab to the lab to the

brain cells in utero. For instance, if you're, if the brain cells are three months old, would they behave almost nearly identical to three month old brain cells in utero? Yeah. So that's another important point because there is different ways to answer that question. By looking into the electrophysiological circuitries is something that we can compare with preterm babies. These are babies born prematurely. And we can measure their brain waves with electroencephalogram, EEGs. And when we compare those brain waves from these premature babies

to the tissue that we made, they are virtually identical. So in the early stages of brain formation, we can recapulate the formation of the circuitries really well. So they do that, but then they reach, after like a year or so, they kind of reach a plateau. And I think this is because,

first, I mean, we are limited on the number of neurons. And second, we for now are lacking input and output. So these brain cells are not receiving any stimulation from the environment. We can do that, but when we measure that, there was no stimulation added. So right now, for the most part, an algorithm wouldn't be able to differentiate between three-month-year-old brain cells grown in the lab and three-month-year-old brain cells in utero theoretically. That's correct.

And the reason why it plateaus after one year is because they're not yet vascularized. They're not getting the blood flow, the nutrients. But that's still, that's not impossible, right? So eventually you could have cells that are older than one, two, three, four, and they continue to evolve similar to a human. Right, right. And that's our goal is to create something that will continue to develop as the human brain does. And I think it's inevitable that we will reach a level,

of complexity of these tissues that would be very similar to the human brain. That's the ultimate goal, is to fully reconstruct the human brain. So these cells are alive. Are they aware? I mean, are they conscious? Are they responsive? Yeah. So we ask ourselves this all the time.

And it's hard because it's always hard to measure consciousness into something that is not similar to you, right? I mean, I can tell you are conscious because you look like me and I can ask you, are you conscious and you can say yes. I can feel my dog or my cat that they are conscious, but as you get something that is not as similar to us becomes hard. So the organoid has no way to express themselves. So I have no idea if they're conscious or not. But we did one experiment to we start probing to that, which is if we treat the organoids with, I have a anesthetics, the brain waves that they form should go away momentaneously. And that's what happens in our brains when we are under anesthesia. Our brain waves stop function. The brain really

calms down and go into a quiescent state until the effect of the anesthesia is wash out. So we repeat this experience with the brain organoids and they reproduce, they mimic exactly what happens in the human brain. So it's not a definitive proof that they have consciousness, but it's an evidence that they are behaving in the same way as the human brain. And so is the science community kind of drying the moral ethical lines of how far their research goes with respect to consciousness, or it's because we can't yet prove consciousness and we don't know how to relate it to matter? So at this point in time, the jury's still out? Yes. We are living in a gray zone. There are some people that believes that they might already have like some level of consciousness.

And so we should put a limit on the research or at least provide guidelines. And there are other people that think that these structures are so reductionist that we shouldn't worry about them and there is no ethical concern. So that's the duality and there is like, again, a gray zone in between where most of the researchers are. as we develop more and more experiments with these organoids, I think we might get to a point where we might conclude that they have some level of consciousness, which would never be the same consciousness that you and I have,

is something else. We'll call it a consciousness of an organoid, and we'll have to decide what that means. I don't think it will block the research. We work with animal models that are conscious, and we just provide guidelines to do it in a more humane way. So if the organoids might be something similar. And arguably, if we can get all of this right, we don't need to work with animal models

that we know and can verify feel pain and are awake and all of that. So this is a potentially more ethical path and a more accurate one. And I remember you stated in a talk, what we can't build, we can't truly understand. And so if we can actually build the brain, we can hopefully understand a lot more. And you use a word, you call it an avatar. So we're building a living brain avatar of somebody in the lab theoretically. Yeah. So you could remodel somebody's neurological development and kind of rewind the clock

and see how their brain evolved and maybe where things went wrong if it is malfunctioning. Yeah. Yeah. So I use the word avatar because we can create these structures from live people. So if I take your skin cells, I can go to the lab, reprogram the cells into these pluripotent stem cells, and from there, it started recreating your brain. So it would be the equivalent of your brain at the embryonic stage

that will evolve in parallel as your brain now. It captures all the genetic structure, all the genome from you, from the person. So if there is like a genetic mutation in there, it will capture as well. And if that genetic mutation is affecting how your brain develops, we'll see it as we grow. And that's why I am applying that to understanding, for example, autism or epilepsy or conditions that happens when you are born and develops throughout life. I mean, when that happens, can we interfere? Are they reversible?

So these are the kind of questions that we are having. So what could your research tell me or tell a patient that's been diagnosed with, a psychiatric disorder or brain-based illness that a psychiatrist couldn't. Yeah. So we could inform, for example, what are the structure or the molecular pathway that was affected? This would be virtually impossible for a psychiatry to do. And then, once we know these molecular pathways, we can find either drugs or a gene therapy or any other therapeutic alternative that might interfere with that pathway, bring it back to a more

neurotypical pathway. So that's what we can do. So we have an information that a neurologist would be impossible to get. So we can review that by doing these experiments. So theoretically, you could take my skin cells, you could revert them into stem cells, grow them into my examples of my own brain organized, my brain cells, and then you could predict, for instance, what neurological diseases I may be at risk of, or if I'm already suffering from what. how that happened and then theoretically test solutions. Could you also test toxins on my brain? So this is how my cells respond to alcohol or to forever chemicals.

Can you see that? All the both, right? I mean, most of the research is focusing on people who already have the disease. So I'll give you the example of my son, who is part of my research. He has like profound autism and epilepsy. And since he was born. So we recreate his brain cells and brain structure in the lab, and we are understanding exactly what happens during embryogenesis

that causes his brain to fire in a different way that is actually now detrimental to him. And we are finding ways to treat him using these information. So that's one thing. You also mention about predicting future diseases. We can do that. But remember, the model that we have is so good that for me to predict how your brain would behave when you are 70, 80 years old,

I'll have to mature that brain for 70, 80 years old, and then you are gone. So right now, there is only one way to speed up the maturation of this tissue in the lab, which is by growing them in space, which is another thing that we do. Wait, I think I have to stop. So you could take my brain cells to speak. and under that stress, they're going to age much more quickly, and therefore you could see, theoretically, me 10 years into the future and age myself that way.

That's exactly what we did. So because we want to use that as a diagnostic tool, we were looking for ways to matured these brain cells in a short period of time, so we don't have to wait. So if we take your brains and age them up to like 80 years old, And if we see signs of Alzheimer's disease, we might tell that, yes, you might be susceptible to Alzheimer's. So that's exactly what we are doing right now. And so does that also mean going to space has a quite negative impact?

Like, microgravity has a negative impact on the human brain, and it stresses it and ages it much more quickly than living on Earth? Yeah, yeah. So that's the conclusion of our research that is space, the space environment, and that includes microgravity. cosmic radiation, magnetic field. We don't know the factor, but in space environment, do age or senes your cells, make them age faster,

even for the astronauts. So the astronauts are more susceptible as well. So this happens not only with brain cells, but all the cells in your body. But most of the tissue can regenerate. So your skin cells after a time in space age as well. But as you get back, your skin cells will regenerate so you look young again, right?

Your blood cells, the same thing will regenerate. But your brain does not regenerate. So the damage that happens in the astronaut brains is likely forever. Fascinating. My goodness. Okay, so theoretically then, if you look at a brain-based cancer, for instance, or a virus like Zika, what can brain organoids then show? us or teach us or how can we treat them to understand those diseases better? Could you reverse or

treat some of those viruses and illnesses using brain organoids? Yeah. And I'll give you two examples. One was with the Zika virus and I was fortunate to have a sample of the Zika virus very early on during the outbreak in Brazil because I'm from Brazil, so I have my connections over there. And we exposed the Zika virus to the organoids and we could see that it kills. some of the cells that makes the brain creating the macrosephalic phenotype that we see in the kids. And by having this essay in the lab, we could test eventual other retrovirus that is already available

that could be repurposed for the Zika virus. And that's exactly what we did. So in two years, which is a record for science, we were able to confirm that the Zika virus was causative to the outbreak in Brazil because it wasn't sure. if it was the virus. And second, to find a treatment. So if there's a new outbreak, we already know how to treat the patients.

So that's one thing. The other thing is, for example, when we applied the same idea to the COVID-19. So we got the coronavirus and we exposed to the brain organoids. And we see, instead of killing the cells, we saw a reduction in the number of glutamatergic. These are excitatory synapses in the cortex. So the interesting point here is that we did that very early on during the pandemic that nobody believed. So when people describe having brain fog and feeling mentally slower after COVID, you can see that on the brain cells in the lab, it is no longer a question or something theoretical.

You can model it on the brain and definitively say, yes, the brain cells or the neurons are firing more slowly or not the same way they did pre-COVID or pre-any virus. or pre-any virus. Yeah, so we predicted the neurological symptoms of COVID because it was so early on and people were still thinking, oh, this is a pulmonary condition, right? It should just affect your lungs. But you're saying, no, it actually affects the brain. And everybody, because we didn't have enough information about these patients to show

that they have a brain fog or psychiatric like disorders, psychosis, things like that. then nobody believed. So we were able to predict what was coming, which is amazing. So we can do that for any emergent viruses in the future. And that brings me to my next question. There was a study done by CAMH, which is one of the leading mental health and addiction research hospitals, and they're based in Canada. And they collected data from 11,000 teens.

And they found that teens who used cannabis were 11 times more likely to develop a psychiatric disorder than teens who don't. So what has exposing neurons in the lab, has that been done to show the impact of smoking weed on the brain? Can we definitively now say that it does have an impact on psychiatric disorders? Yeah, we could eventually do that. Actually, we expose the human brain to cannabidial, CBD. We never tested all the different cannaminoids that is in weed. So we are isolating all the different ones.

And in CBD, even a single dose, doing development, meaning that when you are so young, it might affect your brain, your networks even month after the exposure. So, yeah, the embryonic exposure, I mean pregnant woman smoking weed might have detrimental consequences to the fetus. And then even if it's a young brain, so somebody is in their teens, they are more susceptible to, if they have it genetically. because I think that you could be predisposed based on your genes to certain psychiatric disorders. I think one was the CNR1 gene, which modulates your risk factor. So could there be a future world where a parent or somebody could predict ahead of time, your child may be more at risk of developing a psychiatric disorder if they are exposed to marijuana or other toxins? We would recommend that they avoid that.

And specifically this child versus the one to the left. So that's perfect. you are combining the genetic information we call pharmacogenetics with the modeling because the pharmacogenetics will give you a percentage. Yeah, you have like a specific variance

that might make you a process weed in a different way and that's why I mean people experience weed in different ways. But this gives you like just a percentage we think it's about 30%. But we don't know. But if you use the brain model,

we can give you certainty. Yeah, this person actually processed in a very different way. And we actually performed this experiment in a clinical trial for autism using cannibal. And we can tell that different kids responded in different way. And it's a combination of their genetics as well as how they process weed in their brains. Yeah, so that's exactly where the science is going. So making so many questions from here. So with your lab being one of the first in the world to be able to grow brain cells that nearly behave identically to the brain cells in a fetus, this gives us a rare window into how a baby's brain wires before birth.

So what has your research shown about the impact of toxins such as alcohol or different forever chemicals on the developing babies? brain. Yeah. So this is one of the works on alcohol, where again we expose a single time, only once the organoids to alcohol. And we saw like several alterations, molecular pathways that were completely damaged. Even the identity of the cells, some of the cells should make what we call astrocytes. These are type of the cells that help the brain to wire. And they are completely affected in the presence of alcohol. So we confirmed that alcohol during pregnancy is bad, but now we know exactly why. So that's the kind of a research that we can do.

And there are other environmental factors that we are exposed that we don't know yet if they are good or bad, for example. We think they're neutral, but they might not be neutral. I'll give you examples. We are studying now the impact of it's a molecule called P-FAS and P-FOS. Forever chemicals. Forever chemicals. And these are in the Teflon.

These are like hydrophobic molecules everywhere in your carpet, in your life. It's a pandemic, truly, of mycopstics and forever chemicals. We are all contaminated with that, right? And so what is the consequence of that? Well, maybe for normal people, there is nothing, but maybe there is a subset of people that are more susceptible. So here it comes. I mean, if you have a genetic susceptibility to a neuroscientific,

psychiatric condition, and now you have what we call a second hit, you are exposed to this forever chemicals. That might change the trajectory of how your brain is wired. And these are the kind of experiments that we can perform in the lab. This is absolutely astounding. So you could, somebody could be genetically more at risk for a certain brain-based disorder, a brain-based illness, and exposure, environmental exposure to something like a forever chemical could impact their brain uniquely. And now you could see exactly who that could be, who that candidate is, and how a forever chemical or an environmental toxin impacts their development. So we may one day be able to connect somebody's development of a neurological condition to the items that they wore or the things that

they were exposed to. And that becomes definitive. Yeah. Yeah. So my goal in life, if I can predict the future, it is for every baby that is born, we should have their full genome-sense. sequence, all the genetic information, but as well as they are mini brains. And maybe we can do like a mini lung, we can do like a mini every tissue, so we can visualize, anticipate what are the conditions that that person might be susceptible and better inform them on their lifestyle. Oh, you should not smoke. You should not have alcohol. You should not expose yourself to certain environmental factors. So that will be the ultimate personalized test. To be a little.

to have that predictive capability is life-changing. And even if it just means simple life decisions, I guess for most parents, and even for me, it's the genetic data and privacy that makes me a little bit nervous about the genome sequencing. And I think on the one that if we can get it right and we secure the data, how much our world and our health would change as a result of it. But right now we have all of these conflicts of interest with insurance and data theft and just knowing that it's secure. But I agree that being the goal and nothing in health and science becoming a mystery with the human body. And especially with the brain. So your research puts you at the very forefront of understanding how neurological diseases may form. And you are a parent of a child

who you describe as having profound autism. Right. So what has your work shown or enhanced or help us clarify when it comes to the evolution of autism in a developing brain. Yeah. So what we are figuring out is that something similar to what the genetic has predicted, there are different subtypes of autism. That's why we call like a spectrum, because, I mean, every individual is very different and they express autism in different ways. And that's a good point, just to clarify, autism now, more in a medical terms, we divide in level one, two, and three. And this is the level of independence that you have or the level of support that you need. The level three is the one that you need support all the time.

That's my son. But people with level one might not be seeking a treatment or even a cure or anything like that. they are more looking for acceptance and inclusion. And these are not the people that we are focusing. We're focusing more on the level three. So these are people that if they left unattendant, they will eventually die, right? So we need full attention to these kids,

and they need magical treatment. And what we are learning is that there are molecular pathways that are common among different subtypes of autism. And that's good news, because it means that if we find a medicine that can help one of them, it might be able to help many of them. And when you say molecular patterns,

and what does that mean physiologically? Or what does that look like? What's a molecular pattern? Yeah. So it means, for example, there are certain metabolism that they process in a different way than neurotypical, normal people would do it. And this could be like too much of a certain thing

or too little of certain thing. And if we can adjust these molecules in their brains, they might be able to recover whatever intellectual disability or epileptic that they might have. It's what we call the balance. The brain needs to be in a homeostasis level to work properly. And autistic individuals might have unbalance on certain metabolites. And that's the things that we are discovering. And so according to the CDC, now about 1 in 31 children are diagnosed with autism.

but the data show most kids are diagnosed after the age of two. So is your research able to show earlier signs of autism in development? Or, I mean, what are you seen in the lab that doctors don't see until the age of two? Yeah, yeah, that's a great point, because why we need to wait for two years of age to finish the diagnostic or at least to have like a better certain about the diagnostic. It's because in the first two years, the person is developing. And there is variability in people. Some people, I mean, start walking and talking very early on.

Other people take a little bit more time. So autism sometimes might be in that mix. And if the person doesn't talk in two years, well, they might talk six months after that. And that will not be like a sign off. autism. But the tools that we have might help doctors to claim that diagnostic very early on as early as several months of age. Again, I mean, we need to prove that this is the case. And the only way to prove it is a prospective science. So we need to create brain organoids from, let's say, a thousand people. And if the ratio of autism is like 1%, we should have like 100 of these

organoids create from people behaving in a very different way. So that will give us the power to actually conclude that the tool as a diagnostic is real. We haven't done that experiment because it's very hard to find funds for this type of predicted medicine. But theoretically, the earlier that you could put that diagnosis on, the better. So there could be a world or maybe close to birth, you're able to see that if you're able to do those experiments and model it out. Yeah. That's why I think as soon as you are born, we should be collecting cells and doing, again, the genetic testing as well as the modeling, the brain modeling test. And at your talk I attended in Austin, you stated that we used to think autism couldn't be reversed, but the science is now showing that's not true.

So have you been able to reverse autism in your lab in the brain cells? Yes, yes. Actually, for all the different subtypes that we tested so far, they are all reversible. So all these alterations that I mentioned to you, either on the metabolism level or the number of synapses or the circuitry, if we understand what's causing, could be like a genetic mutation. And if we fix that, it's all gone. So the autism could literally be reversible. That's correct, yeah. And so where does the research need to go?

I mean, I'm sure that anybody listening to this that hears this. is wondering, why doesn't this we should have more funding? Why isn't this the priority? That's my question, too. I mean, I think we should add more funds to that. That is one thing in academia that we call the Valley of Death. And it is in between the scientific discovery and the translation of that discovery to people.

So because you need to run clinical trials. So we find pharmacological interventions, gene therapy ideas that can reverse autism in the lab. But then, I mean, we need to do the clinical trials in people to see if they will actually work. And there are reasons for them not to work, and then we go back again and try something else, but we need to start testing them.

So we have lots of ideas to be tested, but clinical trials takes time and costs a lot. And usually the NIH, or the most funding agents, here in US doesn't support that part. So we depend on pharmaceuticals or industry to kind of a jump in debt and start supporting these clinical trials, again, because the cost is super high.

So that's what we call the Valley of Death. It is usually where the discovery could not be translated back into people. It's all funding issue, no. And is so is it a gene therapy that you work with the brain cells on to reverse autism? Is that what you're actually doing? or is it almost like a genetic surgery?

Yeah, so we are agnostic about the treatment, right? Sometimes it's just a chemical. It's a medicine, it's a pill that would take and revert that. Sometimes it's a gene therapy. So you know exactly the genetic causes, and you go there, and you do either you fix the mutation, or you just replace the gene by the correct version. The mutated gene, you just replaced by the correct version.

all the tests that we perform in the lab shows that this is enough for the neurons to start firing in a neurotypical way. And do you know where the mutation comes from at this point? So you're probably born with it. So it's something that could maybe be tested. Yeah. Most of the autistic individuals are born with those mutations. So it happens during the embryogenesis or before. An embryogenesis is the formation of the embryo.

Right, right. Or it comes and remember that, I mean, each person carries the DNA for, from mom and dad, right? And sometimes these mutations might not do anything for mom, but in a combination with the genome from the dad, or vice versa, it creates the problem, right? It affects how the cells behave. We can figure this out by sequencing the genome of the person.

And in some cases, we do find the mutation. We know, oh, that's the gene that's mutated. Let's fix it. And in some cases, we don't. Either because we don't fully understand the genetics of autism or because it's what we call a multifactorial. There are different genetic variants and we cannot pinpoint just one. What about with Alzheimer's?

Because if you're, so you're taking brain cells, for example, to the International Space Station, you're able to accelerate aging. What does it, does the research show about being able to reverse some of those states, if that's possible? So the research is now in a point where we are convinced. We're advancing ourselves that we can model Alzheimer's with these two. And by sending the organoids to the International Space Station and bring them back, we are seeing signs of deterioration, signs that are normally seen in dementia.

And this includes like inflammation, neurodegeneration. So all those factors that are associated with Alzheimer's, dementia, late-onset diseases, we are seeing the organoids returning from the space station. The next step, it is either a neuroprotection, can we protect those brain cells before they go there, or can we treat them after they come back? And that's my Amazon project. So I'm testing molecules from the plants of the Amazonian biodiversity to see if one of these molecules can do one of those or both, can be a neuroprotector or can be curative. And I know you're from Brazil, but what was the instinct that the Amazon may have a potential cure for Alzheimer's? Yeah.

So I start interacting with different tribes in the Amazon. These are regional tribes that have very few contact with the external world. They live in there. They have their lifestyle in there. And they are now, I mean, having more and more contact with people from outside. And one of my colleagues who are from the University of Manals, who is in the middle of the Amazon, introduced me to some of these shamans. And these are the old people over there who has the knowledge of the mad scene of the forest.

And during my interactions, I always ask, so why do you do when people have a seizure? Or what do you do when people start forgetting things? and it's incredible that they have an answer for every single of these questions. And when I ask about Alzheimer's, of course they have no idea what is Alzheimer's, because most of their population lives up to 100, and they don't show any signs of dementia. And I found that amazing. How come? And they said, yeah, if people start showing like they are forgetting things,

there is this plant that we combine with this other one, and we create like a tea or a brew, and they start incorporating that into the diet. And then I start asking, what are those plants? Can you point it to me? And they point, oh, is this one and that one? And I start realizing that we never studied them. So the modern science completely ignore this ancient knowledge.

And I said, no, this cannot be true. I mean, people must have studied those plants, but no, there's no track record. And in the Amazon, we have about 200,000 species in there, and we probably are aware of 1%. All the rest is completely ignore, but those tribes understand them. So I start recurating those plans and trying to understand what are the molecules that might be neuroprotectant.

And we have, like, a few of them that are good candidates that we want to test into the International Space Station, either again as a neuroprotectant or a treatment for Alzheimer's and dementias and perhaps any of the same. other neurodegeneration. So not only could we potentially build an Alzheimer-proof society, but we could maybe reverse it, and this cure might already just be grown in a plant in nature. It is fascinating, yes.

And it's here the whole time, just to the left of us. And so I imagine you're not going to the Amazon to take all the plants. You're going to recreate those molecules in the lab. So the plants stay preserved in the Amazon. And then what happens to the tribe? Do they kind of share in the research? Did they share in the gains and the winds? We reach an agreement where all eventual royalties of a future product.

If there is a future product, because science has a risk, it might be that we could not reproduce any of those, right? But let's suppose that, yeah, we do find something that is neuroprotected. And if these ever become like a commercial product, riots of those, we'll go back to the protection of those tribes. So there is a percentage of the royalties that belongs to them. And maybe it's incentive for people to not keep burning and attacking Amazon. Yeah.

Because all of the cures in the future could be right within those trees. And they understand that very clearly. It is us that are not aware of that. And so I must ask, with the human brain, I mean, we are an incredibly complex species compared to others. not superior, just different and more complex. And that's why I think we play piano, we go to space, we contemplate the origins of the universe. But is there an evolutionary downside to the complexities of our human brain?

Are we paying some evolutionary cost to be this cognitively superior in some ways? Yeah, so that's another question of the lab. That's actually how I started my lab asking those questions. So I was way more interest on the evolution of the lab. the human brain than on treating diseases. It was my son who actually changed my directions, but part of my lab continues to study evolution. And we have those questions. I mean, what makes the human brain so unique? Why we as any species are so different from the other ones, why we change the environment while, I mean, other species just enjoy them. And I initially

start comparing the human brain with a chimpanzee, because the chimpanzee is a chimpanzee is a the closest evolutionary relative that is alive compared to us. But soon I realized that we are so different from a chimpanzee that is not worth it, because any of these discovery cannot really tell me exactly how the human brain evolved. And then I thought, okay, I need to compare the modern human brain with extinct humans that are no longer here.

Neanderthals? Such as the Neanderthals. And so we start looking at the genome of the Neanderthals because we cannot have cells from the Neanderthals to reprogram and create their mini brains, their organoid. There is no way to do that. But we have the genomic information. And then, I mean, we did like a very simple experiment.

So we tried to contrast the genetic information, the genome sequence from the Neanderthals with the human population, the modern human population. And we try to incorporate. as much a diversity as possible. And then we align them, and they are incredibly similar, right? Very similar. But then we ask the question, is there any genes in there that are different between them

and us? And what is unique about us that they don't have? And by the way, no other species will have, just us. And we end up with a list of 61 genes. These are genes that we all have, the Neanderthals have, all the other species have. But there are sequences that are unique to modern humans. And one of them calls my attention

because it's a master regulator of neurodevelopment. And again, looking into the brain as a tool to understand the complexity of cognition, I decided to swap the Neanderthal version of the gene for the modern human in one of our brain organoids. And what happens was amazing because I thought that, well, I might not see anything because it might be multifatorial. But that single mutation in that gene, just the Neanderthal version in that gene, make the brain organoid mature 10 times faster than normal. So it will take, let me explain that.

It will take a couple of weeks for the neurons to mature and start firing, as I pointed out, in the early stages of neurodevelopment. Now, if I have, like, identical neurons, except for the Neanderthal version, those neurons will mature way faster. Because they're more simple or not as complex? No, because that gene, the master regulator, changes how the maturation of the brain behaves. So it's not regarding the complexity, but it's the timing. And initially, when we had that data, it sounds like counterintuitive. intuitive. But thinking more, what we are figuring out is that the Neanderthal brain develops much more closer to a chimpanzee than us. Because the chimpanzee also developed faster.

That was one of the problems that I had. They could never match the development of neurons from the chimpanzee in modern humans. And the Neanderthals are clustering with the chimpanzee. So they are developing faster. And actually, if you observe a baby chimp, it's way smarter than a human brain. Does that mean that if the brain develops faster for Neanderthal or Chippan Z, they're smarter earlier? Yeah. But so why didn't they go off to then build the Eiffel Tower?

Right, right, because it plateaus. So they develop early because the brain is wired for survival. So they have to go into the wild and figure out how. to survive, right? So they develop quite early, but then they plateau in terms of complexity. But the modern human brain, it takes a while to develop.

So look at that. I mean, we are born after nine months of incubation. And even after that, we cannot leave our babies in the wild. So we still need to feed them for like several years until they become independent. So the developmental time is much slower in humans. And that single mutation is helping us to develop slower to achieve a higher complexity later in life.

But that higher complexity, there's a cost. And that's the evolutionary trade-off. We don't see autism in chimpanzee. We don't see Alzheimer's in chimpanzee. So the slow development and the complexity of the human brain make us susceptible to these neurological conditions that are human specific. And what if the brain was bigger? So if we could have bigger brains,

would we not have to trade as much complexity for size? Could we have more complex brains, but maybe not as many downsides with neurological disorders? Is it because that everything's so dense? It could be. We don't know. That's an interesting hypothesis. We'll have to test that.

But remember that sometimes it's not linear. Sometimes bigger brains, I mean, whales have bigger brains than us. So it doesn't translate exactly into behavior, complexity. Okay. And then my final question before we jump fully to the future, to my world of foresight, regenerative medicine, it does point to a future where we'll be able to potentially regrow organs or parts of damaged organs in the lab and maybe even in the body eventually. So for somebody who is diagnosed with something like ALS or a spinal cord injury, Is there a future where we could regrow and repair their own cells in the lab or maybe even in their own bodies and perhaps reverse some of those conditions or cure them entirely?

Yeah, I think the answer is yes. And I'll give you like an example that it's a research that we published a couple years ago. We did the experiment in mice because we cannot do it in humans, right? So we remove in a surgery the visual cortex of the animal. So for you to see something, your eyes, your retina with all the photoreceptors, need to capture the information through light, passes through the thalamus, which is a region that connects the brain to the eye, and goes to the cortex where you store your memories, right?

So that's how we see things. So if you remove the visual cortex from your brain, you stop seeing. Even though your eyes are normal, right? It's just you don't finish the circuitry because you don't have a visual cortex. to process that information. So we removed the visual cortex from the animal, from the mice, and the mice stop seeing. And then we replaced that with a brain organoid.

After a while, we put the brain organoid from humans in there, and we let the organoids to kind of accommodate into the brain, and surprisingly enough, they start making connections with the circuitry of the animal, and they become the visual cortex of the mice. So those animals now start seeing again, but they are using the human neurons to visualize things. So it's a chimera, a mouse with some human brain cells that allow them to see again. Yes, yes. So this idea of regenerative medicine, you could extrapolate that, for example, take Parkinson disease,

where you have like a region deep in your brain called this triatum that produces dopaminergic neurons. so these neurons respond to dopamine, and in Parkinson, these neurons degenerate. So it could eventually recreate from that same person, these triatone in the lab, and transplanted back and restore the movement of the person with Parkinson's. You could cure Parkinson's by doing this kind of a transplantation. And how far away, I know it's impossible to predict the future in terms of timelines,

but curing something like Parkinson's, and that the model exists, the theory is there, how far away do you think we could go from lab to human trials? Yeah. So unfortunately, science goes very slow. And as I mentioned to you, the value of that. So this funding, so someone, a company or industry or a hospital must be interested on that to perform the clinical trials. There are clinical trials for Parkinson's just transplanting dopaminergic neurons. But this was before we had the technology to create two-dimensional tissues. So now, we have to redo these experiments with a more powerful technology.

But it's coming. Once we have proof of principle, and that's why we still use animals for research. We have to show proof of principle in an animal, and then we can reproduce that in humans. It's a slow transition between the preclinical through the clinical side. And so in some ways we're moving out of the chemical era, where we would treat or cure or at least maintain and stabilize conditions using chemicals, pharmaceuticals, and into the biological era, where we go back to what evolution gave us, tinker it a little bit, and cure with the body's own tools. That's correct. Yeah. If we can do that, it usually is better. Because for us

to fully understand how, again, the disease works, the process, and to find the right chemical, the right pharmaceutical, to actually correct the defect or the molecular pathway that it's affected, it takes time. But the cells, knows how to do that. So if you can just transplant the cells back, let them do the magic. Yeah. Okay. So now I want to jump a little bit into the world of foresight and to see where things could potentially be going. When we look at the field of artificial intelligence and the AI systems we have today, they're really impressive, but they still aren't as flexible or as reliable, even as the brain of a cat or a dog, right? And that's because evolution has billions of years

on AI. Yeah. But scientists are exploring what happens if we do merge evolution with artificial intelligence. And even if you look at AI, I mean the amount of power and water that it uses. And so what happens if we power AI systems with brain cells? So you get the benefits of biology and we're super efficient, but the superpowers of AI. So the field is organoid intelligence and your workplaces you at the forefront. Where is the science today?

And where is it going? Yeah. So we are on the initial phases of that, trying to understand what is, for example, the computational power of an organoid. And we start by asking very simple questions. Can an organoid learn something? Can an organoid memorize something? Can it retrieves the memory? Basically, we are asking, can it have the cognition of a human even if it's not as complex? And we do that by stimulating the organoids. electrical impulses, for example, and we give them different impulses and see if they remember one of them. And how would you show that a brain cell in a lab remembered something? Yeah, so the experiment is like that. So you choose a specific impulse, let's say, certain frequency, and certain intensity. And then you train the organoid to respond to that. So you give that every minute and you see what kind of a response as it has. And we can map that. We can map these networks, right?

And then you start giving like random sequence to see if they would respond in the same way. And guess what? They do not. So they have a specific response for specific impulses, which is how we respond to impulses as well. The way I hold like a rose and a hammer is very different because I treat them in different ways.

And the organoid does the same. Different impulses, they respond in different ways. So then we let the organoid rest, rest for 24 hours. And then we start giving random impulses again, and they will have like their noise in there. And then in the middle of that, we add the right impulses that we know how they responded. And guess what? They remember. And you don't need to even give all the complete impulses.

You just started. They are, I know, it's coming. Yeah. And they responded in the same way. So that's how we proved that they have memory. So we can prove that they have memory. So does that mean that they can remember the same way you could train an AI system? So how are you training AI and powering it on these brain cells?

What does that even entail? Yeah. So that's another thing that we are beginning to explore, which is this concept of generalization, right? I mean, to train an AI, you need lots of training. But the human brain doesn't need lots of training, right? a human baby will figure out that this is a wall. And all the time when it sees a wall, never would beat the wall again

because you're right if you get out. So it generalizes. So how the human brain does that? It's a mystery. We don't know. And we are asking if the organoids can do the same. And the way we are doing is by creating like an interface with a robotic machine. This is a robot that has like four legs.

And we are using the electrical activity of that organoid to make the robot move. And we say the electrical activity, that's when a brain cell is wiring and firing, it produces some activity. That's correct. And that is going to power the robot. That's correct. Yes. Yes.

And then we did one step further, which is to add the sensor information into the robot. So the sensor in the robot that we have is kind of an infrared that will detect when it's getting closer to a wall. And by getting closer to a wall, when it's about 10 centimeters before it hits the wall, it stimulates the organoid with that impulses, specific frequencies that we determine predetermined. And the organoid responded to that because, I mean, it's already known that they have to respond to that.

Then we use that response to turn the robot to left or right. Okay? So that's the training. And now we are exposing this robotic platform to a maze and seeing if the organoid is able to make the robot navigate that maze with different configurations just with one training.

And apparently it does. So a robot powered by human brain cells is able to navigate a maze, even though the cells aren't actually seen it. Yes. So there is a potential future where AI is powered by human brain cells. Yeah, yeah. And that means that the water crisis, the energy crisis that AI currently occupies are dissolved in an instant.

That's correct. Yeah. We could solve the major AI problems by getting inspiration from the organic intelligence that already exist. That, again, evolution took like millions of years to build that. Compared to an organoid, any artificial intelligence algorithm, any artificial network is an insult. Right? I mean, we are not even close to what nature can do. and that's what the kind of power that we want to leverage.

And most likely this will happen in two waves. The first wave would be using the organoid as a black box. We don't understand, but we know it has the power to compute. That's what we are doing right now. But the next step is when we learn exactly how to recreate, reconstruct the circuit trees that do that, something that we don't know. And once we were able to do that,

maybe we can create novel algorithms that mimics that. So we don't longer need the organoids. We can just use organic-inspire algorithms. Right, because I was going to ask, I mean, I know we don't understand consciousness enough to prove, disprove it, so people sometimes say we can't disprove or prove that current AI systems are conscious.

I don't really entertain that as much. I'm not as concerned about a silicone chip being conscious, but an AI system powered on human brain cells, that's slightly different. Yeah. And we're potentially giving it memories, giving it all of this data, and that is a conversation I do have.

So is there a future where these cells could become very sophisticated because they are trained on the same data that AI is? And we see some awareness in AI because it is powered on biology. I share with you, I'm not to worry about current AI reaching a consciousness level

but if we start using the algorithms coming from the organic cells, I think it's going to be inevitable. Because the brain... Wait, so you just said it will be inevitable that these AI systems will be conscious. That's correct. Yeah. Yeah.

That's what the brain is wired for. The brain is wired to become conscious. And if we start using them to power AI, they will inevitably become conscious. Okay. And... Should we cut? Okay.

If the brain evolved to regulate the body and we're going to be building these potentially complex conscious AI systems and then we would give them or embody them in robotics, then we technically have walking, talking conscious robots.

Yeah. Yeah. That's correct. Yeah. We are moving into a cyborg. or a replicant, like just to quote a blade runner, right? Yeah, so you're getting to that stage. Yeah.

And what are the ethical and moral lines the science community has kind of drawn? Because I feel like there's the tech community that thinks maybe we're building consciousness in these silicone chips. And then there's the science community that's like, ha ha, jokes on you. We are actually potentially building conscious, could build conscious AI systems in the lab. So what is kind of the self-policing, if any, or is it? kind of just mutually understood, let's maybe not do this yet until we understand it a bit better? That's usually that's a good idea. Yeah.

And remember that I mentioned that there is two phases. Yes. The first phase is really the biology. And I'm not to worry about that phase. Creating like an organoid that's conscious. I don't care because, I mean, we are conscious. And you can always build like a way to destroy the system. But once we pass to the next.

next phase, which is the algorithm. So then there is no way back. So we have to decide that if we are moving to the next stage. And yeah, we have to, as humanity, to think if that's what we want to do. Building actual things that feel and think and experience. Yes. Yeah. But then the question is, how do we do it safely and how do we do ethically? Because the precedent also is, I mean, even IVF, we were so close to not having it. Right. because of all of the moral panic that it caused. And now millions and millions of babies have been born via IVF. A lot of people go through the egg freezing process.

And that was something that almost didn't happen because we didn't understand the ethical lines and we didn't understand, you know, the idea of creating and tampering with life in new places is something that we immediately jumped to as out of our hands. Yeah. But I guess technology, it does change. our ethics over time.

And for the most part, it's been a good thing. And so I guess the question becomes, how do we see some of these potentially great areas, creating conscious sentient systems being a huge one? And how do we bound that? What decisions do we want to make about it? And I think what's also fascinating about this research and knowing that we are probably going here is we have time.

right? And that's why a lot of people, we end up panicking about the future. But the reality is the future, it doesn't happen suddenly. We see it in the labs. We know we can see it in the data. And so the question is, how do we do things ahead of time properly with more voices weighing in? So it doesn't feel like we suddenly wake up to the future and have to panic about it. And I think that this is one of those moments where if the scientists are saying potentially building conscious machines, is going to be possible, then this is where the humanity steps in and says, what boundaries do we want to put around that? And I think it's the combination of the two that allows us to get it right. And we've done that historically, from IVF to editing to gene therapy treatments, came from historically more controversial science experiments. So perhaps this is one of those moments. And even simple things like blood transfusion, which was weird in the beginning. But now, I mean, yeah, normal. Most people would accept that.

Organ transplantation was another one. Seems like a controversial in the beginning. Oh, I'm going to have like a heart from somebody else. But now, yes, if you don't do that, you die. So maybe these, if you don't do that, you die. It might be the answer. But I like your point that we have time. I don't know how much.

But we do have some time to start raising awareness that these new technology might come. Yeah. And it might take. like up to, I don't know, 30, 50 years, but we will eventually come. Yeah. Yeah. Tools have always been extensions of us, right?

So fire changed how the brain developed. And wasn't fire responsible for us becoming much more sophisticated as a species because we could process our nutrients better? Right. And we could move energy to the brain versus digestion. And so our smartphones in some ways are also an extension of us, right? I think Google Maps is kind of your hippocampus and it's doing the navigation. And then you have your visual memories.

And then the next step, is AI. So you'll have your personal AI assistant and has a fiduciary duty to you and it works on your behalf and it doesn't just write your emails for you but it's maybe solving scientific problems in a startup that you want to bring to life.

Is there a world in which your personal AI system is maybe also powered by your personal brain cells and so it's really a second brain? Yeah, yeah, I think so. That's this idea that you can first of all you can create an organoid from a person

and we discuss a lot about disease, right, try to figure out disease. But it might become, yeah, your new person, your new brain. And the part that I like the most is that these organoid, these brain cells will be created in wire dictated by your genetics, true. But the input to create memories, experiences, it's all about you. So it's not like a clone of you, a clone of your brain, but it's something that has your genetics.

My might respond in a different way. So, yeah, I think it's a nice use of the technology. Okay, so that one's possible. Yeah. And I'm optimistic. I think it might be like a good thing. Yeah, yeah. And so my final question for you is your research doesn't just explore how the brain is built

and to try to understand it and to reconstruct it. But what would a world look like if we could build better brains? So what does that mean more senses, better memory, how could we build a better brain? I avoid using the term better. Okay. Right? So we are already there. We are creating different brains.

Different brains. Yeah. And I'll give you an example of one of the projects that we have. The human brain cannot sense magnetism. But we used to. We used to. We lost that during evolution. And by magnetism, you mean how whales and sharks.

Exactly. It's the fear of your magnet, they understand the magnetic waves of the world. If you look back into our genome, the genes that gives the whale, the ability to sense magnetism is in our genome. But because we never used, or whatever reason, evolution mutated that, so it's no longer functional. Okay? So we are going back and reconstructing those genes, make those genes work back again. So then we're going to have like an organoid that's able to sense magnetism. And what benefit would that give us?

If we were humans that could then also sense magnetism, we don't need Google Maps. We can just... That's correct, yes, yes. So in that experiment was done with blind people. So they put like blind people in a forest to see if they will figure out a way to get out. They couldn't. Again, because we don't have magnetism. But if our brain now has the sense of magnetism, yes.

I was more thinking, I mean, the reason for this work is to, better create sonners, right? I was not thinking about augmentation of the human brain, but that's definitely a possibility. I'll give you like another example. This is our work with NASA. Since we figured out that the astronaut brains are susceptible to the damage coming from the space environment,

we are wondering if this is caused by cosmic radiation. And most likely, there is a contribution for cosmic radiation. Can we create a human brain? that is protected by cosmic radiation. So what we did was to clone a gene that's coming from the tardigrade. I don't know if you... Kind of the organism? Yeah.

Okay, so for somebody, again, who missed that biology class... If you miss the biology class, tardigrade are microscopic pairs or entities that is everywhere. And they are incredibly resistant, resistant to fire, to cold. So that's why they dominate the... planet. They are everywhere. And we end up contaminating even the space station and even the moon. So there are now tardigrates in the moon because we took them there. Right. So we are actually responsible for all the alien conspiracy theories. We are actually sending them out words. We are already contaminating the universe. Yes. So in the outside of the space station,

there are tardigrates. So the question is how they survive the cosmic radiation. So we now know there is a gene in their genome called this soup that suppresses the mutation causes by cosmic radiation. So what we did was to clone that gene inside the genome of a human and create an organoid that now is resistant to radiation. So we are doing tests at the space station to see if that organoid will work in the same way as the tardigrade would be resistant to cosmic radiation. So if it's positive, well, we might think about engineer the human genome for future astronauts

that will go in missions, interplanetary missions. So that might be like something to consider. Especially as space becomes the next frontier, we need to make sure we're resilient to some of the adversarial effect of it. Yeah, yeah. So yeah, yeah. The combination of genetics and stem cells are really powerful. It makes you dream.

It makes you dream and brings it to life. Well, it has been an absolute pleasure. This has been so fascinating, and I can't wait to do it again. Super. Thank you so much. Thanks for coming. Thanks for joining us for this episode of I've Got Questions. If you've got questions about AI and emerging technologies, send us a message or a voice note on our website.

And if you enjoyed this episode, please like and subscribe and share with someone you think might be interested. We'll see you next time.