In this Author‑to‑Author episode, Tim Trott delivers a sweeping, narrative‑driven survey of artificial intelligence from its mythic origins to contemporary practice and future challenges. He begins with ancient Greek myths of bronze automata such as Talos, moves through classical philosophy (Plato, Aristotle) and early mechanical inventions across cultures, and then outlines the mathematical foundations laid by Boole, Frege, Ada Lovelace, Pascal, and Babbage. Trott explains Alan Turing’s universal machine and how it sparked the idea that intelligence could be formalized. He recounts the 1956 Dartmouth conference that birthed AI as a field, the early symbolic‑rule‑based systems (SHRDLU, expert systems), their limitations, and the subsequent AI winters. The narrative then shifts to the resurgence of neural networks, the back‑propagation breakthrough, and the deep‑learning era powered by big data and GPUs, highlighting modern applications in voice assistants, autonomous vehicles, healthcare, finance, and logistics. Trott concludes with a discussion of ethical and societal concerns—bias, accountability, regulation, job displacement—and emerging research directions such as explainable AI, human‑AI collaboration, quantum‑AI hybrids, and climate‑focused AI. Throughout, he intersperses sponsorship for Cybershoot web hosting and a call to action for authors. The episode frames AI history as a continuous human fascination with creating artificial life, urging viewers to consider not just what AI can do, but what we will choose to do with it.
Chapters
This is Author to Author, and I'm Tim Trott. Author to Author is sponsored by Cybershoot website hosting, where your author website needs to be. Today we begin a series of programs of special interest to authors. The title for this chapter is "History of Artificial Intelligence." The history of artificial intelligence, from Greek myths to modern technology. Yeah, I said Greek myths. That's where we're starting because people have been dreaming about artificial intelligence for thousands of years. They just didn't call it that. Ancient stuff first, myths, philosophy, the first mechanical devices, then the math and the machines that laid the groundwork. The birth of AI as an actual field arrived in the '50s, followed by the emergence of large language models and chat engines. So where does the idea of artificial intelligence even come from? Well, it starts way back, thousands of years before computers. We're going to look at the ancient Greek myths about mechanical beings made of bronze, but somehow alive. Then we'll talk about what early philosophers thought intelligence actually was and how the mind works. And finally, we'll see the real machines people were building— automata, mechanical devices in ancient and medieval times that actually could move and act on their own. The ancient Greeks loved stories about artificial beings. One of the most famous is Talos, a giant bronze warrior who supposedly protected the island of Crete. According to the myth, he was built by Hephaestus, the god of the blacksmiths and craftsmen. Talos would patrol the shores throwing boulders at enemy ships. He was mechanical, not born, but he moved and thought on his own. There were other stories too, mechanical servants, golden maidens that could walk and talk, all these automata that the gods supposedly created. Now obviously the Greeks couldn't actually build anything like that, but that's not the point. These stories show that people have been dreaming about artificial life for a very long time. They were trying to imagine what it would mean to create something that could actually move, think, and act independently, inviting the question: What is intelligence? What makes something alive? Can you build it? These myths became archetypes, patterns that show up over and over in how we think about AI. The created being, the servant who might rebel. The artificial guardian. We're still telling the same stories today, just with robots and computers instead of bronze statues. The questions haven't changed. We've just gotten closer to being able to answer them. Way back in ancient Greece, philosophers were already asking the big questions. What is intelligence, really? How do we know things? How does reasoning work? Plato and Aristotle spent a lot of time thinking about that. Aristotle in particular did something really important. He came up with a formal logic, basically a set of rules for how to think correctly. If A is true and B follows from A, then B must be true. Sounds obvious, but writing it down as a system, that was huge. And 2,000 years later, when people started trying to program computers to think, they went right back to Aristotle's logic. It was the first instruction manual for reasoning. But here's what these early philosophers really gave us. They forced people to ask hard questions like, what's the difference between a human thinking and a really complicated abacus following rules? Is there even a difference? Where does consciousness fit in? Can you have knowledge without understanding? Those questions about what makes human intelligence special, those are still the questions we're wrestling with in AI today. The Greeks didn't answer them, but they made sure we'd keep asking. For thousands of years, people have been trying to build machines that move on their own. Ancient Greeks built mechanical theaters with moving figures. Chinese inventors created water-powered devices that could pour wine or strike bells at a specific time. Islamic scholars during the medieval period designed elaborate mechanical servants, singing birds, and fountains that changed patterns. These things ran on water power, steam, clockwork gears, whatever they could get to work. Some of the machines actually did useful work, but honestly, a lot of them were just for show, to impress people or entertain royalty. And that's fine. They still proved something important. They showed that you could capture motion, capture behavior, and recreate it mechanically. What's interesting is this wasn't just about practicality. People were trying to bring stories to life, to make myths real. Those ancient tales about mechanical servants and artificial beings, people wanted to actually build them. And slowly, over centuries, they got better at it. That's the thread that runs from ancient automata all the way to modern robots and AI. This human obsession with creating artificial life and making machines that can do what we do. So how do we get from ancient philosophers to actual thinking machines? Three big steps. First, mathematicians figured out how to express logic and reasoning as actual math problems. Then inventors started building physical machines, calculators, mechanical devices that could do things automatically. And then Turing came along with the idea that tied it all together. The 1800s brought some huge breakthroughs in how we think about thinking. First, you had George Boole, who came up with Boolean algebra, basically a way to express logic as math. True or false, yes or no, on and off. Sounds simple, but it's the foundation of how every computer makes decisions today. Then there was Gottlob Frege, who took it even further with formal logic. He created a mathematical language for expressing any logical argument. It was rigorous and precise, but it was exactly what you'd need to teach a machine how to reason. And here's someone who doesn't get enough credit. Ada Lovelace. She was working with Charles Babbage on that analytical engine we mentioned before. But she saw something nobody else did. She realized this machine wasn't just a fancy calculator. It could be programmed to do anything you could describe in logical steps. She literally wrote this in the 1840s. That's right, a woman was the first person to really grasp what programmable computation meant. So what did all this add up to? These people proved that thinking, at least logical mathematical thinking, could be broken down into steps, and if you could break it down into steps, you can mechanize it. That's the idea that eventually became computer science, and later, AI. Centuries before computers, people were already trying to build machines that could think, or at least calculate. In the 1600s, a French mathematician named Blaise Pascal built one of the first mechanical calculators. It could add and subtract using gears and wheels. Pretty impressive for that time. Then in the 1800s, you had Charles Babbage, who designed something way more ambitious, the Analytical Engine. This thing was supposed to be programmable. You could give it different instructions and it would solve different problems. Sounds obvious now, but back then that was revolutionary. The problem was, it was so complex that He never actually finished building it. But here's why these machines matter. They proved something important: you could automate thinking, at least the mathematical kind. You didn't need a human to sit there and do calculations. A machine could follow rules and make decisions on its own. That idea, that machines could be programmed to solve problems, is the foundation on which everything else is built. No Pascal and Babbage, no computers. No computers, no AI. Back in 1936, Alan Turing had his brilliant idea. He imagined a simple machine, just a tape that could read and write symbols following a set of rules. Sounds basic, right? But here's the thing: Turing proved that this simple machine could, in theory, solve any problem that could be solved step by step. Any problem at all. That was huge. For the first time, people had a clear answer to the question, what can a machine actually compute? It laid the groundwork for everything in computer science today. But it did something else too. Turing's work got people thinking. If a machine can follow any set of instructions, and if human thinking is just following instructions in our brains, well, couldn't a machine think? That's the question that eventually became AI. Can we break down intelligence into steps a machine can follow? That's what Turing made people wonder. And when programmable computers were built a few years later, they were built with his ideas in mind. So let's talk about how AI became a field of study. We'll start with a conference at Dartmouth and the people who were there. Then we'll get into how they approached the problem, all that symbol manipulation and rule-based logic. And yeah, there were early wins and failures, and there were plenty of both. The field of AI really got its start at that conference at Dartmouth College in 1956. Some of the biggest names in the field were there: John McCarthy, who coined the term artificial intelligence, Marvin Minsky, Alan Newell. These guys were incredibly optimistic. They thought, we can build machines that can think like humans. And they made progress. They built early programs that could solve logic puzzles and play games. Nothing too complicated, but it was a start. The thing is, their goals were huge, maybe too huge. They genuinely believed that they were close to replicating human intelligence. But what that conference really did was make AI legitimate. It brought researchers together, got universities interested, and attracted funding. It created a community of people all working on the same big problem. That foundation, the institutions, the academic programs, the network of researchers, That's what made everything else possible. In the early days, AI was about symbols and rules. Researchers would try to break down human thinking into step-by-step logic. If this, then that. They'd feed computers tons of rules and let them crunch through the problems. There were some pretty impressive systems built that way. In the '70s, there was one called SHRDLU. That could understand natural language commands and manipulate virtual blocks. Then you had expert systems, programs loaded with knowledge from specialists in fields like medicine and engineering. And within their narrow areas, they worked, sort of. But there was a catch. These systems were a nightmare to scale up. Every new situation needed new rules. And the real world? The real world is messy and ambiguous. Things don't fit into neat categories. A symbol-based system would just fall apart when faced with that kind of uncertainty. So eventually researchers realized, okay, we need to try something completely different. And that's when they started looking at different approaches. Early on, AI researchers had some real wins. They built programs that could prove mathematical theorems and play games like chess. Remember those? People got excited, maybe too excited. The problem was that Researchers kept promising way more than they could deliver. "We'll have human-level AI in 10 years," they said. And then nothing. Investors got burned, governments cut funding, scientists outside the field started rolling their eyes whenever anyone mentioned AI. Progress basically ground to a halt. Twice, actually. They called those periods "AI winters." And yeah, it was pretty bleak. But here's the thing: those failures taught everyone an important lesson. The human brain is ridiculously complicated, and building something that thinks like us is way harder than anyone thought. Duh! So, researchers stopped trying to do everything at once. Instead, they explored different approaches, different techniques, and different ways of tackling the problem. And that diversity of ideas, that's actually what eventually led to breakthroughs. So, how did we get from those early experiments to AI? 3 big things happened. First, neural networks made a comeback and got way more sophisticated. Then deep learning took off. That's when researchers started stacking these networks dozens of layers deep, and they got some incredible results. And finally, all that research started showing up in real products. Your phone understanding what you say, your car's camera recognizing a stop sign, your bank software deciding whether to approve a loan. And that's all this technology at work. Scientists looked at how the human brain works and thought, what if we could build something like that? So they created neural networks, basically artificial neurons that connect to each other and learn from information that's fed in. Back in the late '50s and early '60s, they built something called a perceptron. It worked, well, sort of. But it could only do the real simple stuff, and people got discouraged in neural networks fell out of favor for a while. Then in the '80s, researchers figured out something called backpropagation, which is basically a way to teach these networks by showing them their mistakes and letting them correct themselves. That was the breakthrough. Suddenly, you could build much bigger, more powerful systems. And that's when neural networks got really good at recognizing patterns, like telling the difference between a cat and a dog in a picture, and sorting things into categories. That's what eventually led to the AI we have now. Then around 2010, AI took a huge leap forward when researchers started using real deep neural networks with tons of layers. What made it finally work was having massive amounts of data to train on, plus powerful graphics cards that could handle all that number crunching. These new setups were built for specific jobs, some for understanding images, others for processing language. And they got really good at it, better than humans at recognizing what's in a picture. Major breakthroughs in getting computers to understand how we actually talk and write. That was the key. And all of a sudden, AI is everywhere. Your phone's voice assistant uses it to understand what you're saying. Self-driving cars can see the road, and doctors can use it to spot things in x-rays and scans. Banks use AI. To make smarter decisions about loans and investments, and shipping companies figure out the fastest delivery routes. Basically, AI went from being this academic research thing to something that actually affects how we live and work every day around the world. There are, however, certain ethical considerations and societal impacts for AI in daily life and industry. Future potential and ongoing research. AI brings up some serious problems, like who's watching what these systems know about us? Are they treating everyone fairly? Are they biased against certain groups? What happens to people whose jobs get replaced by machines? And when AI makes a bad decision, who is responsible? Researchers and government officials keep saying we need AI that people can actually understand and trust, something that's fair and doesn't cause harm. Right now, there's a lot of arguing about how to regulate that, and how to protect people's rights, and who should be in charge of making the rules. Everyone's trying to figure out how we can keep moving forward with AI without letting it run wild and cause damage. AI is pretty much built into everything now. When Amazon suggests products you might like, that's AI. When a factory knows a machine's about to break down before it actually does, that's AI. Business analysts say AI is making companies more productive across the board, and they expect it to keep growing. Mm, sure, all these AI tools make life easier, but they also mean we all have to learn new ways of doing things. Customers have to get used to dealing with chatbots instead of real people. Businesses have to completely rethink how they operate. It's convenient, sure, but it's also a lot to keep up with. So what are researchers focusing on now? They're trying to make AI that can actually explain why it's made a decision instead of just being a black box. They want to build AI that's more like human intelligence, not just good at one specific task. They're figuring out how humans and AI can work together. They want to make sure AI is safe and won't do things that we don't want it to do. Some exciting new stuff is happening too. Scientists are starting to combine AI with quantum computers. Which could make everything way faster. And people are using AI to tackle climate change, figuring out how to use less energy, predict weather patterns, and so on. Now, will there be an AI that's as smart as humans at everything? Some people think we'll see it in 20 or 30 years, but most experts say let's not get ahead of ourselves. Progress will be step by step, not some sudden breakthrough. And they all agree that scientists, ethicists, policymakers, and regular people need to work to make sure AI develops in a way that helps humanity instead of causing more problems. We've covered a lot, from ancient stories about mechanical servants all the way to the AI that we're dealing with now. We looked at how machine learning works, and we talked about the real problems that AI is creating in our society. Knowing where AI came from helps us understand why it matters so much now and why we need to think hard about the ethical questions it raises. Here's the thing: going forward, the big question isn't "What can AI do?" We know it can do a lot. The real question is, "What are we going to do with it?" We need to stay curious about how it works. We need to think critically about what we're being told AI can and should do. And we need to make sure that as AI gets more powerful, it reflects what we actually care about as humans, and not just what's technically possible. Or profitable. And that's something to think about. From one author to another, your readers deserve a real home to find you. That's why I built Cybershoot Hosting. For $10 a month, you get secure WordPress hosting, backups, email, security of SSL, a firewall, and a free author theme with a setup included. No bots, no scripts, just real human support and USA servers. Unlike social media or Goodreads, your own site lets you sell direct and track your marketing with a Facebook pixel. Make your author platform truly your own with Cybershoot Hosting. And there's a free bonus if you follow this link. Disponible en español. Thank you for being here. Please share this video with your friends on social media. Be sure to click the like and subscribe buttons and click the bell so you'll know when to join us for the next chapter of Author to Author. Author.
