Welcome to CXO Talk episode 860. I'm Michael Krigsman, and we are discussing AI and the future of work with Doctor Terry Signowski, a pioneer in computational neuroscience and deep learning. He's the Francis Crick Chair at the Salk Institute for Biological Studies and is also a distinguished professor in the Department of Neurobiology at UC San Diego. He is also Chairman of the Nur IPS Foundation, which runs the largest, most prestigious AI
conference in the world. Terry has published over 500 scientific papers and 12 books, including his most recent, ChatGPT and The Future of AI, and has won more awards than I can name. I have a background in physics, which is basically a license to work on anything, but my real focus is on the brain. I've won awards in neuroscience, the most recently the Brain prize, which is the highest
award. And you know, I was a pioneer in the early days in the 80s developing learning algorithms for neural networks with Jeff Hinton and others. And so I have the, these two different backgrounds, which I've now have reached the point where this is really what's exciting is they're coming together. They're, they're actually synergistic. And So what we learned in the brain is going to help better AI in the future, and what we learned in AI is actually
helping us in neuroscience. So this is really an exciting time to be doing research. How do you model AI on the brain? Maybe we should start there. Back in the 20th century, artificial intelligence was really focused on things that ran on puny digital computers, namely symbols, logic and rules. And there's only so far you can go with that, especially when you're dealing with a, a complex, very complex world with lots of uncertainties in it,
right? And so you need, you need a much, much more computation and, and, and, and now we know, in fact, you know, that nature has solved all these problems, vision, you know, coordination, motor and so forth. And so we're learning now we popped the hood and now we're learning. What do you take from the brain? Massively parallel, highly interconnected parameters that called synapses in the brain and learning that you learn the strengths of those synapses with experience.
And, and now it's big data and that has transformed all of AI And, but the point is though, we're dealing with the same mathematical structures, right? These neural networks are very simplified versions of brains, but they have the same mathematical way of of of analyzing and making progress. How does the brain organize information differently from
machines and neural networks? And I just want to mention that the prominent data scientist Anthony Scrifignano has helped me prepare some of these questions. The general principles I've, I've given you take you so far, but actually nature has evolved of specialized circuits for specialized problems in, in different species in order for survival. And you know the, the different, there are many, many differences.
There is the, the deep learning architecture that is used today was inspired by the cortex, the cerebral cortex, which is the highest level on top of the brain that humans use for, you know, the knowledge base. And there's a hundred other brain areas that are equally important for her being able to handle the complexities of the world and to survive. And, and a lot of those are ones that are right now are being integrated into the, the, the next generation of AI.
And I'll just give you one example, reinforcement learning. So reinforcement learning is, is the part of the brain that is, is found in almost all species and has to do with being able to, first of all, from experience, predict whether a particular action is going to give you a reward or not. And you, you build that up over many, many trials. You know, if you're wrong, you twiddle awaits.
And that's actually used for what's called procedural learning, learning how to play tennis, for example, or learning how to do us, you know, solve problems that are, you know, in law, in medicine, all of that is, is done. It becomes automatic. And, and, and now that's being incorporated into AI and it's, it's going to help, you know, advance it and, and make it more, much more broad and much more reliable. Can you drill into this a little
bit? How does how do you make that transference of understanding the brain into encoding this into a digital system? It's all about architecture. That is to say, you have the same units, but you could put them together in different layers and with different connectivity. And you know, in Transformers, which is the, the, the, the heart of large language models, there's, there's special things that were added in order to be able to make it a, a, a powerful language learner.
And by the way, that it was trained completely on the basis of just predicting the next word in a sentence, right? It, it wasn't given any special training on any special tasks. But now that we trained it up on a huge database that can solve many, many natural language problems that before were very difficult and then required specialized networks.
And so that's one step closer to the fact that we know we have a very general, we, our brains are very general in terms of being able to handle lots of different problems. But OK, so reinforcement learning in particular is, is really, I, I think a part of the brain that is absolutely essential. It's all about practice. If you want to learn how to play tennis, there's two options. You can read books about tennis, right? That's the declarative part, the cortical part, the conscious
part. Or you could go out on the tennis court and start hitting them and getting feedback and getting better and better day after day, week after week. And, and that eventually automatizes it so that you don't have to think about it anymore. And, and, but of course, that means you can't explain it to anybody. So if someone asks you, how did you do the serve? Well, you put your hand up and you hit the ball right. And that does it really help them?
What about explain ability? Can you discuss the challenges of explainability that arise from neural networks and machine learning? And this has become very important and in business and and Justin. Let me first point something out that's obvious, which is that everybody has a brain and uses it to solve problems and doesn't know how it does that right. The brain is not explainable is that it doesn't give us insight to how we solve the problem, right? But, and we make mistakes, we
hallucinate. Actually, we we make things up, filling in the blanks, right? We can't remember in detail what happened, you know, a couple years ago, but we can kind of fill it in. So, you know, this is not, you know, it's unusual that we're creating a system that is like the brain in some ways that it's also going to have some of the problems of the brain. So OK, but now let's look at
explainability. Explainability is not just one level of understanding and, and let's take medicine, right? Well, you go to the doctor, you get a pill and the doctor will tell you, oh, it's going to help your immune system, you know, and maybe that's good enough for the patient, but it's actually not good enough if you're, you know, trying to do research on medicines and try to understand, you know, how it works. And that's another level explanation.
But, you know, most, most people would be, you know, you know, don't have the background to understand that you have to have, you know, be an expert. But even that, OK, there's a lot of going on in the brain that that even experts don't understand. It's just really complicated, You know, all of the different molecules interacting and
signals going back and forth. You know, we, we have a very crude understanding of, of really how, how nature put together the, all the different organs in the body and how they interact with each other. We're discovering that the, the gut actually has a connectivity, you know, and send signals back and forth with the brain, the gut, you know, the microbiome, right? This is something we didn't know until recently and we're learning things every day now.
But you can go even down deeper. You can go down and understanding, you know, the actual details of the molecules and, and this is physics explanation, right? This is something that is at the level where you have equations and, and that level of explanation. All these explanations, you know, are, are, have different ways of, of giving us insights and, and you know, the, the one that we're looking for. Is it words?
No, I think we're, we're really, we really want to get an understanding of what's happening at that deepest level. And that's beginning to happen because we have a lot of very, very good mathematicians and people, engineers with the training and machine learning are digging down into these networks and figuring out this the actual way that they are, you know, how the information flows through the network and how how it's able to do the amazing things that it does.
And and, you know, this is really where we're going to get the real understanding, not just in terms of some explanation that, you know, you could live, the explanation that you might use just to make you think that you've understood something. As we rely more and more on AI to make decisions, including medical decisions, you, you just alluded to that this issue of explainability will become more and more important.
So the real question for from that I have is can we ever really understand how the machine is making these decisions? And then what do we do about the fact that we that we're asking machines to ultimately decide important aspects of our lives and we don't know fully what's going on? You shouldn't think of AI as being a human. It's not right. It's it has special capabilities. It can, it can take in much more knowledge than any single human,
right. So it's, it's really in some way already super intelligent, although people argue that it's not intelligent, but nonetheless it's helpful. So AI is a tool. It's going to help you solve problems and, and you talk about medical of decisions that have to be made. So a study was done. It was published in nature. Nature is the gold standard for science, all sciences and engineering.
And, and what they did was a study on lesion skin lesions, some of which are cancerous and some of which are benign, right? And it's like 2000. And what they did was they compared experts or panel of experts with the best AI solution. AI was given thousands and thousands of examples, images of of lesions and it was taught to classify them. Now it turned out that both groups had a performance of about 90%.
They were correct in telling you what kind of lesion or whether or not it was cancerous at about the same level. OK, so that's interesting. However, if they let the doctor use the AI in making the diagnosis, the performance went up to 98%. How could that be? I mean, they, they started with 90. Well, the reason is that they have different knowledge. The AI had access to much, much more data than any individual could ever see in their lifetime in terms of different rare skin
conditions. And the doctor had much deeper knowledge about the the ones that are the most common. So, you know, and and it's a Doctor Who's calling the shots, but the doctor takes advantage of this other partner in order to make better diagnosis. And so this is what what's laying out now in every area where it's being used by writers, by people who have to, you know, work on ad copy, people who have, you know, engineers are trying to solve
problem. By the way, the by far the most impact that a is had over the last 10 years has been in science has transformed science, because science is filled with big data on specific projects that are really very complex, you know, big particle accelerators, huge, huge amounts of data that's being generated. And, and it's it's really transformed the way scientists do science. And so this is really a harbinger of what to expect in
other areas. Please subscribe to our newsletter, go to cxotalk.com, subscribe to our newsletter. And we do have some questions coming in. So let's jump to LinkedIn and we have a question from Greg Walters who said he read some somewhere that AGILLMSAI, digital twins, etcetera visualize actions and processes like athletes and and more a million hours of study in one second. Is this correct? And how does this visualization impact model collapse and does
it? This gets to the heart of of how you actually create one of these large language models. You start with a huge amount of data, trillions of words, and you go through this process of, of, of predicting the next word and gradually getting to the point where you can with very high accuracy. And the only way you could have gotten that where that point is by being able to disambiguate. Words have multiple meanings and you have to look at the context.
And so it has, it has to create an internal representation of the semantics, the meaning of the sentence, in order to be able to understand just the way you do it, right? You can predict the next word, but you know, you have to understand what the sentence means before you can do that. So but that takes a huge amount of time and and very expensive. The the the latest GPT 4, for example, took two months with 20,000 GP US of graphics processing units and cost $100
million, right. So, but the point is that not only now can the, can the, the large language model chap GPT answer new questions, but it can do that like in you press the button and you get the answer in, you know, one or two seconds, right? And, and that's shocking. I, when I first did it, I just couldn't believe this is how could this be? And the reason is it's internalized all this information so that it immediately is able to respond literally, you know, without
actually having to think. In fact, these large, they don't think, they don't think the way humans think. And that's one of the things we're going to try to improve. The next generation is going to actually have internal self generated activity. Right now, if you stop your dialogue with ChatGPT, it goes blank. It doesn't keep thinking about things the way that you do.
If you're on your own without any sensory input, You think about planning, you think about what happened, you know, in the past. Chat GDP doesn't do that. It doesn't have an internal life. This is from Arsalan Khan and on Twitter. And he says artificial intelligence is a mirror of human biases. How can AI and our brain become less biased quickly? And I I think this leads us directly into a discussion of the impact of AI on our workforce and work displacements and all those implications.
The bias could come in because the data are, are incomplete or skewed. And, and, and, and this came out, for example, when face recognition was being used. It was much, much, much more accurate with white faces and black faces. And it turns out that there were many fewer black faces that were used in a training set. So you go back to the training set, you have to curate it.
You have to put, make sure there's, and now we're discovering it. The higher the quality of the data, the better the performance in general. But there's other things you can do. And this is, this is a little bit more tricky right now. You know, you give, you have to give a goal because it is, you know, deep learning networks don't have intrinsic goals the way that humans have goals. So you have to give it the goal. And the goal typically is I want to optimize performance on a
particular class of problems. And so you give it a lot of examples and it, it chugs away and, and it gets better and better and better and better. And the larger and larger the network, the, the better and better it gets. But the point though, is that it, it may not be aligned with a lot of other values that you have, like fairness, right? And So what do you do? Well, you're going to have to add fairness as one of the goals that it has.
And now you have a problem. OK, I want maximum performance, but I also want maximum fairness. I can't have both, right? So I have to wait them. I have to decide, well, half and half or maybe 90% performance and 10% fairness, you know, that that's a decision has to be made explicitly. And you know, we're somehow
we're making it implicitly. And when we deal with the world, you know, given the cultural biases that we have now, here's something about bias, though, and, and I just want to put this to you, OK, I'm going to ask a question of you and your audience, OK? You know, humans are definitely biased, right? We see this all time and LO, Ms. reflect that to some extent because they're trained on human
data. Here's the question, which do you think are going to be easier to fix, the LLM or the human? I would. Assume the human is easier to fix because with the human you have control in quotes over one, whereas to fix bias in the LLM you need to adjust a very large data set. Actually, I'd be curious to know what your audience thinks. Folks, you heard the question. In my experience, I've known a lot of different people and very rare that they change their minds about it.
They're really fixed in their ways. You know, as you get older, I mean, maybe younger children, OK, there, there, there's some hope there of, of changing their by. But you come into the world, you don't know what the language is. You don't know what the values
are of the culture. And you learn through reinforcement learning, by the way, you know, what's, what's things that you, that you should, how, how to deal with relationships, how, how to value, you know, you know, people's advice, think things that you know, that are different in different cultures, how you deal with, you know, truth and so forth. It's it's different, it's not the same and and you learn. And Greg Walters on LinkedIn
agrees with you. He comes back and he said it's the LLM that will be easier to. That's because we engineered it. We can fix it. But how? How do you fix that bias given there are so many levels of insidious? OK, OK, OK, so he he in my book, you know, chat GDP and the future of I, I actually laid out because I there's a there's a lot in the book that is is is
really inspired by the brain. And by the way, there's a revolution going on in neuroscience that is equally important as what's happening in AI. And this has to do with the brain initiative, Obama's brain initiative. It's been 10 years on and we have made enormous strides in understanding the mechanisms underlying your brain. And so we're, we're going to actually be able to really, you know, get, get, come up again with a much better explanation for the decisions that we make.
But, but, but here, here's, you know, when, when we're dealing with, as I told you with for humans, you use reinforcement learning. So here's what has to happen in AI instead of just training the network at the beginning. And then that's it. You, what you're going to have to do is use reinforcement learning all along. We need lifelong learning where we have reinforcement learning going hand in hand with the, the, the declarative, it's called the declarative knowledge system.
And, and so we have to basically treat the LLM like a child, right, You know, and, and punish the child when it does say something wrong, does something wrong, and that how else is it going to learn that right? We have to align it with our values. And, and, and that's beginning to happen. And I know a couple of companies already used it, not for this particular purpose, but for example, something called chain of thought.
When you want to solve a problem, you break into pieces and you solve each one separately. Mathematicians do this all time when they're proving a theorem. But the LLMS, if you ask them, they just jump to the answer and sometimes they're right, sometimes they're wrong. But now you can actually train using reinforcement learning, this habit of breaking into pieces and solving each one separately. And now it's gone. The performance has gone on these mathematical problems from
20% to 80%. So this is an example, you know, you're using reinforcement learning. It's actually in part a, a way of a way of solving problems and actually a method that could help not not just solve mathematical problems, but any problem that requires, you know, in, in in manufacturing, in figuring out how to reconfigure your office, all of that. Should we should be able to help. We should be able to explain, you know, through a sequence of actions, how you're going to get
to some goal. And in answer to your question to the audience of which will be easier to remove or address bias in the LLM or in people, Isaac Sacolic comes back and he says it's easier to fix AI than humans. People don't change belief systems easily, and data and weights can change in an AI. The real issue is understanding what values the data scientists embedded in their training. So he's in agreement with you. And Arsalan Khan is also in agreement. Everybody's against what I said.
Arsalan Khan says LLMS might be easier to change since the input is from multiple people, whereas people rarely changed. Exactly your point. But now let's talk about the implications of all of this on work, the workforce, jobs. Thoughts on that? We are at the stage that the Wright brothers were at the beginning of aviation 100 years ago. The very first flight by the Wright brothers was 10 feet up and 100 feet forward. That was it. They were inspired by how birds
glide by the airfoil. They were inspired by the lightness of the wings, and that was why they use canvas. But the most difficult part of making aviation safe was figuring out how to regulate, you know, and how to control the direction you're going and, and regulate it so that you don't crash. That was the difficult problem. Well, we're, we're living through that right now, except that maybe it won't take 100 years. It'll only take, you know, 1020 years.
So we're, we're, we're, we're just at the very beginning of that process. And what's going to happen over the next 10 years near term is incremental advances and that's already happening. You see all these language models, there's now a dozens of them out there with different capabilities and, you know, focused on different problems
that have to be solved. So, you know, we're, we're just exploring that and, and one of the directions that we're going is actually not bigger and bigger because we've run out of data, but smaller and smaller with quality data, with data that's more specific to your particular company, to your
particular profession. And that now is, is going to shift the balance away from these big high tech companies toward, you know, the enterprise, the companies, the ones that are actually doing, you know, the, the, by far the most important job in society, which was, you know, making things and getting things to work and, and dealing with the complexity of the world. And that, that could be done. This, this is something that is, is, is, is not going to take
just five years. It's not going to take 10 years because it takes is you have to complete reorganization of the way that the, the, your office and your company is organized. So right now, for example, I'll just give one example. You know, you have databases all over the place, especially if you're an international company and you know, how do you
integrate across databases? Well, I mean, you have an IT person who has ways of doing that and, but you know, to answer your question, that requires a lot of comparison and, and, you know, accretion of, of, of, of different sources. That's, that's going to take a long time. But it one of the things that LMS are really good at it because it's already done that for the database of the world. Why not for your company? All of those databases could be integrated.
And now when you want to answer a question, you don't go to the IT guy. You ask your, your ChatGPT. Well, you'd call it something else, right? You know, it would be called chat IBM, you know, and, and that way you're, you're going to be able to have access to instantly to all kinds of important questions that you can answer that may help you make decisions. This is if you're, you know, of course, a head of a company. What about if you're in the middle of the company?
Now, one of the things to keep in mind is that these tools require training. You know, chat GDP out-of-the-box. You can ask the questions, but the answers you get back are often not the most useful. That's because like any tool, you have to learn how to use the tool. And you know, there's a, a whole group of people now that call themselves prompt engineers. And what, what, what does that mean?
It means that they know how, how to use the prompt in order to be able to more quickly get useful answers out. And, and in my book, I go into this in quite some detail that may help, that may help them. And, and the other thing is, is it's, it's really interesting. It's the, you know, if you really want to incentivize people, you, you know, you what you have to do, dealing with chatty P is fine. People do it on their own. People just doing it because
it's a lot of fun, right? Well, once you, once they get trained on doing this, this is going to help with productivity. It's, it's, it's already doing that in the many areas, like I gave the example of medicine, but it, it's, it's going to help people in ways that nobody ever even imagined, right? I mean, this is, it's, it's, this is something that like, you know, what was the killer app for personal computers, right?
It, it wasn't, you know, simply typing and, and you know, the letters, it was Lotus 123. Do you remember that? It was a spreadsheet and, and that suddenly that was something that's very useful that could be used by not just individuals, but by companies and organizing all the data and so forth.
And, and that has become central to the, to the way that now we all deal with large data sets as we, we have them in these super, you know, you know, the spreadsheets now are, are, are much better design than Louis 123. But the same thing is going on right now is that we're, we're, we're beginning to develop the tools that are specific and in surprising ways. Let me give you one example of something that surprised me. It's in my, and again, it's in
my book. So a lot of people have mental problems, you know, they have anxiety, they have phobias and so forth. And, and you know, and, and worse and mental disorders, but you know that what do they do? They go to a psychiatrist or a cognitive therapist and you know, they, they have to make their appointment weeks ahead of time. And then, you know, they go and then, you know, they have an hour and then they get a big
bill right now. You know, this is, this is, you know, it, it helps, it really, it turns out cognitive therapy is as good as taking a pill. It really is. It's and, and it's complementary, by the way. Now they did a test. This is a scientific study. If people are given the choice between an AI psychiatrist or a human psychiatrist, what would they choose? What do you think, Michael? What do you think? I will say that they choose the AI over the psychiatrist you'd.
Figure this out from my first trick question. Yeah you're absolutely right. But majority of, of humans feel more comfortable talking about, you know, personal things that it might be a little embarrassing, right? If they're talking to another human who are, you know, humans are very judgmental, right? I mean, but the AI, you know, it, it's not going to judge you away. It doesn't have any internal care, but you know, it's going to give you respond in a way that is objective and so forth.
So you know and you know this is this already happened back at MIT in the early days of AI when someone put a very simple program that it did very simple thing that just repeated the question. Eliza. Eliza And, you know, a Weisbaum Weisenbaum discovered that his secretary is using it during the lunch break. You know, it's like, and you know, it didn't, she didn't care that it was a machine as in a very simple minded machine. You know, she did it because it kind of helped.
It got her to think about things, you know, explicitly. So there you go. I mean, that's that's unexpected, right? There's a lot of unintended consequences. And by the way, not all good. There's going to be bad ones too. And so that's why we have to really be careful about, you know, what, what to expect. You have just been describing what I would call incremental changes of efficiency, right?
We now have a critical thinking or an information partner to help us improve our writing to research and so forth. But what about the more that the deeper implications and more foundational changes that this may drive in society and in in the workplace? Any any thoughts on that? In the short room, you're not going to lose your job. Your job is going to change and it's going to change because you have these new tools. You're going to have to learn
new skills. And that's why it's really important that we have within companies and, and it turns out that we have a, a ways of doing that that could help companies. This is called massive Open online courses and they're free and there's a lot of out there now there's like, you know, 10,000 and I have one myself called learning how to learn.
And this is with Barbara Oakley. And what we realized was a lot of people, especially 25 to 35, half of them are college educated, are in the workforce and now they need new skills. And it's harder to learn when you're in that position because you can't go back to school and you have mortgage and children. And so you know, you want to be able to be more efficient. We teach you how to be more efficient through what we know about the brain, how the brain learns.
In any case, they're there, but there are all kinds of mooks out there about how to use the new tools of and the idea I think is for people now who have a job are going to be using your tools more efficiently. But gradually the job is going to morph. It's it's going to be the boundaries between jobs may change in terms of, you know, what people can do and integrating more often rather than having stacked layers and so forth. You know, it's, it's impossible to predict the future.
Let me ask you, OK, Michael, you were around in the 90s, right? Yes. OK. You remember when the Internet went public, right? Yeah. Could you have predicted what influence the Internet would have on every aspect of your life today? Absolutely not. Not at that time. OK. Well, we have another technology that's that's going to have a equally important impact up the road that we can't predict. It's, it's just, it has, it has so many different effects, so
many different parts of society. But Terry, I did not invent or discover important aspects, parts of the that underlie the Internet, but you did when it comes to deep learning. And so therefore I look to you as the the the prophet and the seer of what may come. The reality is that I'm actually probably no more capable than you are. In fact, maybe less capable of of predicting the future. You know, the Internet pioneers back then. Let me give you an example, OK. You know, Google, when it
started had a motto, do no evil. Do you remember that? Yeah, absolutely. And, and, you know, that they were very idealistic. They said this is going to democratize information. Everybody is going to have access to information and their voices will be heard. OK, well, could they have predicted how that will play out? And here we are now we're we're dealing with, you know, misinformation, fake news, echo chambers. I mean, this is like, you know, who could have predicted that, right?
I mean, this is something that even the experts didn't predict. The enthusiasm of young people before they became billionaires. That's right. That's right. So I'm a young, I was once a good person in any case. Yeah, we, we, we, we are, you know, I'm, I'm actually an optimist. I, I actually think that we will get it sorted out, but I know it's not going to be easy and it's going to take a long time. And and that's true of all technologies. It's not like it, this is
different. It's just, it's just that it's so new that we really are just beginning to, you know, you know, trying to beginning to understand the capabilities it's moving target, it's incrementally improving there. There may be a breakthrough and it may not be, it may be, you know, I don't know, 10 years, maybe 20 years. And I think it's going to be the breakthrough, I think is going to be when you put something which call is called System 2 into these large language models.
System 1 is what we have right now, which is basically just trained to, to produce, you know, generate the generative AI one word after the next. But this, this system too, is the one I alluded to earlier, which is cell generated activity. And, and we, we actually know how that's done in the human brain. And so it's just going to be a matter of time before that's transplanted into AI. You know it, but it's not going to take place in, in, in 5 or 10 years.
We're talking here for probably, you know, decades out. Chris Peterson on Twitter says the current hype based AI takes vastly more resource inputs per question than a human brain takes for 24 hours of doing everything. Can you talk about the sustainability, the power requirements, the resource requirements? You alluded to this earlier in your discussion of the trade off between completeness versus rooting out bias and the input cost and time associated with that.
We have to set priorities. It can't things have been the cost of things have is gone through the roof and with that cannot continue. And I also alluded to the fact that the next generation they're going to be smaller language models with higher quality data. That's One Direction. The other direction is hardware and I have old chapter on this in my book, right? If you're interested go, you
should go there. So the human brain is able to function at very high levels with 20 watts of power, some a little brighter than others, right? But you know that's much less than the the gigawatts that are out there being used right now to answer your questions. So there's clearly a huge, huge imbalance. And that's because the technology that nature uses is, is based, is gone down to the molecular level.
It's really, really advanced. But you know, we're, we're actually, there's a hundreds of companies out there that are building more efficient technology that is going to make lower power much more the, the architecture that you need for these deep learning networks is completely different from the von Neumann architecture that is in your PC. That architecture does 1 instruction at a time and it's a has to fetch information from memory, which is separate.
Well, in your brain, the memory and the processors are one in the same. They're, they're integrated and there are a lot more processors, right? This is all the neurons. You have 100 billion neurons in your brain. And now we're up to, you know, it, it, you know, literally 100 trillion of synapses in your brain, we're up to a trillion weights in these networks. So we're still like only 1% of the brain even just in raw
power. But here's the beauty is that we can special purpose chips that could be much more efficient and, and and that will take us down by a couple orders of magnitude over the next 10 years already going on. And this has happened to GP us are a good example. That is A2 orders of magnitude when they put GPT, when they transplanted these original deep learning networks onto GPUs, it was tours of magnitude more efficient.
And the reason is that they have thousands of processing units just like all the neurons working in parallel exchanging information. So that was that was a big step. That was a huge jump. And that there's an inflection. The the Moore's law went from doubling every two years to doubling every two months, literally. I mean, this was a huge, huge change. Now that's that's going to happen again.
And it's, it's, it's, it's going to become probably another two orders of magnitude or three, possibly with a, a new class of computing called neuromorphic engineering. What is that? Well, it turns out that if if you take your digital chip, which are basically 0 ones, right, and run them near 0 where things are analog, by the way, digital chips are actually analog when you look at the actual transistors, but they run into the rail.
But if you work down there, and this is Carver B, that Caltech actually understood this, that you can actually take advantage of that processing at that low level for doing crude multiplies and adds much more efficiently by by like two or three orders of magnitude. And now that's mature technology. And so that's going to come online over the next 10 years. So see, what's going to happen is just like, you know, Moore's laws, that the technology gets
more and more efficient. And, and we're, we're just at the beginning of that because we, you know, we realize now that this architecture may be more important than, than the von Neumann architecture for solving many, many kinds of problems that before, you know, we couldn't. This is from Isaac Sukolic and Isaac is a big time CIO influencers. A lot of CIOs listen to him and he says he liked your healthcare example.
It may be hard to benchmark human versus human plus AI versus AI and other diagnostic areas. What are options to validate accuracy and build trust with AI recommendations? I think that's a really important question. That's productivity and, and, and that's the bottom line actually. You know, for a long time, computers were you the, the, the whole business world and, and, and, and public are investing in digital computers and productivity didn't change very
much. It wasn't until you connected them together that they could exchange information. That's when things took off. Let me go back to just answer the question in a concrete way. Let's go back to healthcare. So, you know, you go to a doctor for, you know, you have some medical problem and you know, you walk into the office, you have 20 minutes doctor sitting there next to a computer and is asking you questions, you know, about your symptoms, about your
previous medical history. What is he doing? He's looking at the computer he's typing in because you got to get all that in the computer because you have to have a record, right? And then, you know, he's not looking at you. And then at the end, you know, he, he, he, he gives you a list of, of drugs that you should take or some advice and so forth. And, and off you go. And you know, you, maybe you get, you remember half of it if you're lucky, right? So here's what's happening right now.
There are 10 companies out there that have realized I could put together a lot of different AI systems to make that process of not just a lot more efficient, but actually much more likely to be of help to the to the patient. So first of all, speech recognition, the doctor doesn't have to look at the computer. He can talk to the patient. It turns out you get a lot of information by looking at the patient, not just, you know, what they're saying.
You can, you can you can tell from their facial expressions, you know, how serious something is. You can look at their face, you know the color of the face. I mean, a good diagnostician use can use that information right and, and jump to, you know, to conclusions that you can't just with a couple of numbers that they are that that you're getting, you know, typing in and OK, now what happens?
So you press a button at the end and much more human interaction and now chat GDP does the heavy lifting. It gives you a summary. It's very good at that doctor looks at it and checks to see what's you know, if there's any problem in terms of the recommendations and so forth. And now you know that goes back and, and the the patient now walks out with actually very detailed summary.
And so now the patient is going to be able to understand much better what needs to be done over the next couple of months, you know, in order for the patient to be able to, you know, improve their, their health. And, and you know, this, this is just you kind of one example of how we have a system right now, which is really disadvantages patients, right? And doctors, by the way, doctors don't just stop because they have to write up notes afterwards and often late into
the night. And now all of that is being done. They could, they can go and they, you know, this is something that it's going to take decades to permeate the medical system. You know, they're very conservative. So it's going to take a long time and it has it'll have flaws and it'll there'll be problems, but ultimately I think that it'll greatly improve healthcare.
Chris Peterson has a question, and very quickly please how does the enormous cost of retraining the big models play into GDPR style right to be forgotten laws? Will AI systems need an exception because it's just not practical to retrain for each request and very quickly? In addition to the initial training, which is the costly part, that is also opportunities later to do something called fine tuning. And what's that mean?
Fine tuning is basically giving it extra data and trying to train it in a, in a gentle way such that it now has access to new information without interfering with the existing database. Now, it, it's not perfect. What happens is that the, the, the large language model gets dumbed down to some extent because it's taking on this extra burden. But that's not nearly as expensive. And that can be done in house, right? That could be done in individual
businesses. You, you have to hire somebody who, who can do that for you. You know, that these are machine learning people. And by the way, the, the, you know, I, when I started, we were doing neural networks, right? And that was in the 80s. And then it'd be overtime Neurips became a machine learning conference. But now it, it, it's come back, you know, full circle. And now it's, it's not called neural networks anywhere. It's called AI, right? You know, and it wasn't because
we called it that. It was because the world was calling it that. But it's really about machine learning. That's the heart of AI today. And it's all about data. Whoever has more data wins in any area. And I'll tell you your company is sitting on enormous amounts of data that's really important for you. And if you can get that into an AI, you will win. Can you talk about how the assumptions made during data curation influence AI outcomes
usually? This is really going to become a very large business data, as I said earlier, data is hard to come by quality data even harder. And and so, you know, how do, how do you, how, how is it that we're going to be able to overcome a lot of the problems? You know, hallucination is a good example. You know, hallucination actually is, is has has advantages. If you're a writer or, you know, ad copy, it's really good at that or poems, but not so good if you're asking for a fact
hallucinates, you're in trouble. So I think what we're going to, we're going to need is another layer on top of the, of the language models, which is, is, is a fact checker basically. And, and will, will take what is coming out and we'll learn how to then look through the database world's database and, and come up with, you know, sources for that particular fact. So again, it may be this isn't really the exciting thing is that right now humans are doing
that. You know, that all these big companies have thousands and thousands of humans that are going through and sorting through getting rid of racist data, trying to, you know, make prevent of these large language models from saying from doing anything that is going to hurt people. You know, that that is being done on a individual basis. And that's very labor intensive.
But you know, we, we should be able to train another generation of large language models to actually do that part of the job of, of, of, of sorting through and flagging things.
And, and, you know, even in the earliest days of, for example, neural networks in the 80s, when I started, they were using these with very, very, at that time, they were very shallow with one layer of pin units to take a slide with cancerous, it was called a Pap smear, you know, the cervical cancer and, and humans would have to look through the every cell and it would take hours and hours and very costly.
Well, you know, neural networks could do that much more quickly, reduce it to 100. And then the humans were going to, you know, use their much more powerful visual system to look at the, the debris and, and, and all the different types of cells and, and, and come up with a much actually thorough and faster way of analyzing the data. And so the idea is we have layers and, and humans are one
of the layers. They could be the layer at the top or they could be in the middle or, you know, anywhere. But the idea though, is that we have different systems that are there for spotting different problems in the data. And, and this is, again, this is where we're at the very beginning of this, this is going to take decades. You know, it, it, it took it literally. Just think of the Wright brothers, right? Look how long it took to get to jet engines, right?
And and jet planes now you know you can go across the world now, but back then, right, you could be be lucky to just go across the countryside. So, on the subject of humans and datas and data, how can we detect and understand the impact of adversaries manipulating data, for example, with misinformation and disinformation? Well, that's already happening. You know, we're social media. This is filled with that, you know, that that's, that's a, a
very, very difficult problem. I don't think it's going to be any easy solution. I, I outlined just now what I think is going to happen with all these layers of, of, of, of filtering. But you know, there's, I think one thing that we should be open to is the fact that every once in a while there's a, a massive a breakthrough. The last one occurred in 2022 when chat GDP was opened up to the world and suddenly everybody realized, oh, my God. Yeah. What hath God wrought?
That was the that was the first message sent over the Telegraph. And we're going to have it. I'm sure we'll have another moment like that. I don't know when, I don't know how, but you know, it's going to happen. It's going to happen, and it's already happened. And I told you in science, it's happened multiple times.
I'll just give you one example. Protein folding, how does it a string of amino acids get folded up to become an enzyme that has a function and that turns out to be a computationally intractable. You can't do it. It just takes too much for computing power and, and most biologists thought it would never be solved, but we've solved it. Alpha fold DeepMind created a transformer that could solve that problem and and it's
transformed biology. This is like a huge why because you can now design enzymes that have much more specific actions and better drugs, better ways of, of being able to make predictions about how different interactions are going to occur between proteins. And this is really a transformation in the whole business of, of creating better healthcare, better drugs. So it's going to happen in many, many areas in ways you can't predict.
What advice do you have for business people given everything we've discussed? Number one, read my book, not my lips. Read the book. Because I talk a lot about these issues that, I mean, I don't really touch the surface of what is in the book. OK #2 the most important thing you should be thinking about is training your workforce with this new technology and take it seriously. It's, it's going to, it's it, this is just the beginning, right?
And, and unless you start now, right, helping your workforce, we use these tools, the new tools and, and it really is that they're tools that if you can use them badly and, and, and very well. And you know, people are trained in that. And it does take time. It takes, you know, maybe 100 hours. That's a lot less than a 10,000 hours it takes to become an expert, right? And #3 download your company
into ALLM. And I'll tell you, this is already being done with the brains of flies and zebrafish. These are brains that have 100,000 neurons. We can download them now into AI and and they have similar behaviors, they have similar activity patterns. You know that's the future. My mind is blown from this conversation. Terry Signowski, thank you so much for taking time to be with us. Oh, I'm, I'm very pleased. Thank you so much for making this opportunity possible for me.
And a huge thank you to everybody who watched. You guys are amazing. You asked such excellent questions. Before you go, please subscribe to our newsletter, go to cxotalk.com, subscribe to our newsletter, subscribe to our YouTube channel, and we have amazing shows coming up. Thank you so much everybody, and I hope you have a great day. Take care.