Unlocking the AI Box: Join the Waitlist Now!

0:00Today on the podcast, I want to take you on a nostalgic trip back in time to the early 2000s when there was another similar to ChatGPT kind of chatbot that came out that a lot of people don't know about. So I want to talk about the story of that, what happened to it, and why I think that's important. So the tool used back then was called Smarter Child. So long before ChatGPT came around, Smarter Child was the most impressive AI that many people had seen. And essentially, this was an instant messaging chatbot that I would say was a bit ahead of its time. So it essentially had the knowledge of an encyclopedia.

0:36It had a bit of wit and it was added as an AIM and MSN buddies list like buddy. So it was actually added to over 30 million people's friends lists worldwide back in the early 2000s. So obviously this thing was incredibly popular. And for many people, it was their first kind of use of artificial intelligence, something that is obviously now used by everyone. So Peter Levitin is a co-founder of ActiveBuddy. That's a parent company behind Smarter Child, the actual chatbot. And he once said, quote, We were offering people something they never had before. When you talk to Smarter Child, it knew who you were when you came back.

1:18It was like your friend and having a computer friend then. And now it's fantastic. So I think by today's standards, Smarter Child would definitely seem very primitive compared to something you see in ChatGPT or Anthropic or any other LLM. but this was back in around 2001 so the chatbot really was obviously quite impressive is impressive enough to actually inspire investors to fund the development of siri which in turn you know paved the way for amazon's alexa and a lot of other ai assistants and this was kind of one of the first steps so i will also say though that not all of smarter child's creatives are creators are all as optimistic about the future of ai robert hoffer that's another co-founder of active buddy he's dubbed the bot father.

2:00He often refers to stories like Frankenstein and the myth of Prometheus, and he expresses concern that perhaps humans have gone too far with technology just because we can. A quote from him, he said, It's wonderful that Smarter Child sort of opened the Pandora's box. Unfortunately, now I feel like I have a certain amount of responsibility to share with the world, the good, the bad, and the ugly. So what set Smarter Child apart, in my opinion, was its personality. Smarter Child was known for its kind of sense of humor that they had kind of programmed into it. And unlike the neural AI bots we see today, Smarter Child would respond with, you know, some wit and sass and jokes to users, and it was, they were essentially trying to push its boundaries.

2:43So if, you know, some kid came and gave it a naughty joke or something, it would ask it, like, do you kiss your mother with the same mouth that you say that with? Like, it would just, like, I don't know, just say, like, funny things. So anyways, nowadays, a bot like Like Snapchat's, you know, my AI essentially would respond to anything inappropriate by saying, sorry, I can't respond to that. But back then, Smarter Child would demand an apology when it was treated poorly or it would refuse to interact until the user apologized. It was like a completely different beast back then. In the words of Peter Levitin, he said, what the AI world is not delivering at this moment is really any personality or any soul.

3:22And I think that that may actually be intentional, right? They don't want people to think that these AIs are sentient, and so they're really trying to sterilize any sort of humor out of them or any sort of personality or anything that people might confuse for sentience because they just feel like, you know, then everyone's going to be worried about them taking over the world when they see how advanced they are if they were actually sentient. So because of that, you're not seeing that. I think Smarter Knowledge had a knowledge base that was really impressive at its time. ActiveBuddy licensed data sets and databases from IMDB, the Weather Channel, the Dewey Decimal System, Elias Sports, Yellow Pages, and Sony.

4:05And this really enabled it to provide a massive wealth of information to the users on demand. So today you see people like ChatGPT and Google Barber, they essentially just scrape the internet and grab all the publicly available data they can through web crawlers. Back then, they would actually just get the licensing to use the databases, which is, you know, probably better. But they were then able to use it on demand and they had, you know, updates on real sports and real weather and real events in real time, which I think is something that we probably could learn from today as we're struggling with ChadGPT's still 2021 cutoff, which is starting to seem kind of archaic when you look at other AI models that don't have that same limitation.

4:44When talking about this, Hoffer, who was one of the co-founders, he said, we sort of showed up like a virus, and when we look back at it, it was quite clear that we were definitely the progenitors of all modern bots, from Siri to Amazon's Alexa, all of the various AIs we see today built around large language models. I think Smarter Child really had an impressive rise. They went from zero to 30 million users in less than six months. That's absolutely insane. I think it became a phenomenon of the early 2000s and the internet. But like all things, its moment in the sun eventually came to an end. This thing eventually did not last forever.

5:22In 2006, Microsoft had bought the company that created Smarter Child, planning to use the technology to automate customer service for Xbox. So funny, like that's the use case of this technology with 30 million users. They're like, sweet, we're going to automate the customer service for one of our product lines. And this is obviously far from what this, you know, quirky chat bot had been used for, had been loved by millions of people for. So in the process, Smarter Child was essentially phased out. And by 2008, it was killed at Microsoft. So I think as we look to the future, some like Levithan don't believe AI will ever truly become sentient, while people like Hoffer, his co-founder, expressed fears about the Pandora's box that we might have opened.

6:05Hoffer says, quote, we're right before the opening or maybe we've just opened Jurassic Park. so I think you know it's definitely uh up for debate where we're at on that spectrum but I think it's really interesting listening and hearing from founders who have essentially started a very similar company that have kind of been pioneers and kick-started this um and of course their technology was taken and killed before it really reached scale the same way we saw chat gbt um well I mean I guess it really did reach scale but it was just kind of killed so I think the really ironic thing here is that it was Microsoft that bought it way back then and then killed it and then Microsoft is the main investor in OpenAI but I don't think they're going to kill this one because it's just too plurific and has permeated into much of popular culture today so I think ChatGPT is here to stay and I'm excited to see the advancements it makes and we'll see if it really does become sentient or not.