“They call me an overnight success—it only took me twenty years.”
You’ve heard that joke. Now apply it to AI.
Except it didn’t take twenty years. It took seventy.
The entire history of artificial intelligence in 5 minutes.
AI & The Art of the Possible — Exploring AI beyond the hype
Hosted by Chance Sassano
Episode 2 – The Overnight Success Moment Full Transcript
Can I tell you the entire history of artificial intelligence in under five minutes without putting you to sleep?
I bet I can, but you tell me. Here we go.
I’m chance. And this is the Art of the Possible
Episode Two, The Overnight Success Moment
In 1950, a British mathematician named Alan Turing asked a simple question, can machines think? 75 years later, we finally have an answer. Early researchers were wildly optimistic.
They thought they’d crack AI in 20 years, maybe less early. AI started with programming computers, using rules if this, then that. If the patient has a fever and a cough, then diagnose the flu. If the stock drops, then sell. It worked kind of, but only for narrow problems. It required humans to write every single rule, thousands of them. It was exhausting. It didn’t scale.
By the 1970s, the hype had died. Funding dried up. AI became a punchline. This was the first ever AI winter.
Everyone gave up.
Well, not everyone.
A few researchers kept going. They had a different idea. What if, instead of programming every rule, we let the computers learn the rules themselves, feed a computer thousands of cat and dog photos. Let it figure out the difference. No human writes the rules. The computer discovers patterns in the data. This was Machine Learning.
And suddenly progress, real progress. Computers could recognize handwritten zip codes, built or spam, recommend movies, but they couldn’t understand language. They couldn’t have a conversation. They couldn’t write a coherent sentence. For that, we needed something bigger.
Understanding human language is hard. I mean really hard. The word bank can mean a financial institution or the side of a river. I’m fine. It can mean you’re okay, or I’m fine. You’re definitely not okay. Context. Nuance. Sarcasm. Researchers worked on these challenges for decades.
This led us to something called Natural Language Processing, teaching computers to understand how humans actually talk. Progress was slow, painfully slow.
Then in 2012. Something changed. Researchers built artificial Neural Networks, computer systems, loosely inspired by the human brain, but they didn’t stop at one layer.
They stacked them in hundreds of layers, deep layers. This was called Deep Learning, and suddenly everything got better. Computers could translate languages, recognize speech, generate text that almost made sense. In 2011, IBM’s Watson wins jeopardy in 2026. Google’s alpha go beats the world champion at Go a game far more complex than chess.
We were getting closer, but we still couldn’t have a real conversation with a computer. Not yet. Then researchers tried something audacious. What if we built a really big neural network, not hundreds of parameters, not thousands, but billions, and what if we trained them on almost everything humans had ever written?
Books, websites, articles, Reddit threads. All of it.
We didn’t teach it grammar. We didn’t program syntax rules. We just showed it patterns, billions and billions of patterns, and something unexpected happened. The model didn’t just learn to predict the next word. It learned to reason. To summarize. To write poetry. Answer questions. To have conversations. Not perfectly but well enough.
This was called a Large Language Model, an LLM.
November, 2022. OpenAI releases ChatGPT to the public.
Five days. 1 million users. Two months. 100 million users.
The world woke up. AI wasn’t coming. It was here.
So what just happened?
Machine learning taught computers to learn from data.
Instead of rules, natural language processing taught them to understand human language.
Deep learning made them better at both and large language models, they put it all together.
But here’s the thing, it’s not magic. It’s 70 years of research, thousands of scientists, two AI winters where everyone walked away, and a handful of people who refused to stop believing.
You know that joke about the musician who says, “They call me an overnight sensation. Yeah. It only took me 20 years to get here.”
AI is the same except it took 70.
So the next time someone says AI came outta nowhere, you’ll know better.
We’re not at the beginning of AI.
We’re at the beginning of what we can do with it.
That’s The Art of the Possible. I’m Chance Sassano. Thanks for listening.
Next episode, Val Kilmer lost his voice to cancer. AI gave it back. Doc holiday rides again, The Huckleberry Moment on the art of the possible.
