What's Hot in Artificial Intelligence

0:00Well, we promised a little discussion about AI and kind of open AI so that you really understand it a little bit better, because artificial intelligence AI is rapidly transforming society by shifting from a specialized tool to an integrated algoritmified, and that's a big word for me, system that influences daily life, economic structures, and industrial operations across all gametes. It's going to affect all of us, and it's moving at breakneck next speed. So I thought I'd reach out to a dear friend and a partner of mine that knows a little bit about this, and I've seen him kind of go through the machinations of being someone that really didn't know anything about AI to becoming kind of like an AI expert. In our next best guest, spend over two decades as a congressional chief of staff, the White Campaign Manager and a political spokesman. He's been involved with me for many years, and you may have even seen him on CNN, Fox News, and court TV, and now he's

1:01using AI and AI tools to transform his business and his life in some ways. His name is Randall Thompson. He's the founder of Pulse Communications and the co-creator of one campaign AI that will be launching soon up. It's a platform bringing professional grade AI-powered campaign tools to candidates as well as campaigns across our country. And Randall recently built, are you ready for this? He recently built a team of six artificial intelligence agents. These are like personal secretaries to do your work on open claw to and he'll explain what open claw is in a moment here, but to transform how he does business and Randall, it is so great to be have you on the show. Hey, thanks for having great to be here. Awesome. So Randall AI has been moving at breakneck speed. What's happened in just the last month that people should be paying really close attention to? Well, I mean, in order to understand the last month, you kind of understand the overall

2:01field, the playing field of what we're dealing with. You have three major companies out there. You have Anthropic, you have Open AI, and you have Google. And they're all in kind of this arms race, right? They're all trying to achieve what they call AGI, which is artificial general intelligence. It's kind of the hypothetical AI system that can understand, learn, and apply knowledge across any intellectual task a human can do. So you kind of do that on a scale of one to 100, right? How close are we getting to AGI? And most of the industry professionals say that we'll probably get their early 2027 into 2028. There's 95% chance we'll hit 100% AGI by then. Elon Musk, you'll see that he talks a lot about it on Twitter. And he says that it's called singularity. Singularity is what happens after AGI arrives. It's the theoretical point where AGI and AI become smart enough to improve itself, which then triggers a runaway cycle of intelligence that game goes so fast that human civilization

3:05fundamentally changes in ways we can't predict or comprehend. And that's what freaks everybody out, right? So we're not there yet. But just within the last 30 days, when you go into chat GPT, or when you go into Claude, or when you go into Gemini, which are the main AI models that people use out there, most people all just use chat GPT. It's one single session. So you open it up, you ask it a question, it gives you an answer, you close it down. It doesn't remember what happened from session to session. And in the last month, there's this thing called OpenClaw that is come in that has an LLM, it's called a long language model, which means now you can have those sessions over and over and over and provide it context about your life, about your goals, about your dreams, about your needs, about your wants, and it doesn't forget. It stays open from session to session to session to session to session and refers back and remembers and learns from itself. And that is the thing that has the AI industry up in arms right now about all the

4:06possibilities that are happening until we get to AGI of how this can be implemented both personally and in business. Listen, you've just talked a little bit about the OpenClaw. So you've built something called OpenClaw, a multi agent AI system, kind of like personal secretaries, I guess, that do work that actually are doing work for you and running your political consulting from. What does that mean in plain English and layman's terms to the normal human being? And we've got about three, four minutes. Yeah, no, I'll go quick. So what it means is basically it becomes your own virtual assistant. Think of all the things that you do over and over again during your day. That's called a workflow. I have to bring this in. I have to decide about it. I have to route it to here, route it to there. I have to create an Excel spreadsheet. I have to write a document. I have to do those kind of things. But it's the same task over and over. So if you create that workflow and you give it rules, it's much like hiring a brand new employee. You onboard a brand new employee, the first month or two, you're going to be very careful with them. You're going to over-explain.

5:09You're going to tell them what to do. And then pretty soon, they learn how that process goes and it becomes automated. I'm doing the exact same thing with just six different AI agents. One's a creative writer, one's a researcher, one's a salesperson, one's a quality control specialist, one is a second brain. So it never forgets anything. And then I have Max, my chief of staff, who runs everything in operations. So you're basically saying that people can have these AI agents that they can give personalities or names to that will actually do their work for them when they're sleeping. And they never get sick. They never get tired. You don't have to pay social security on them, correct? You're absolutely right. It's fascinating and it's amazing and it's wonderful and it's also a little scary. Scary. But it's not something that you just can build in 30 minutes. It takes a long time and you pretty much are in the early days of this. You need to find somebody who can invest the amount of time to be able to build an AI agent for you.

6:13But once you have it, I'm telling you, it's amazing. So how do people get in touch with you and learn a little bit more and that boy, you've got a large listening audience. So how do people find out about this or how do people contact somebody that can help them kind of learn about how to set up their own agents if they want them? Well, certainly one of the best things that was ever advice that was given to me was help or use AI to help you learn about AI. Ask them questions. Ask about OpenClaw. Ask about what that means. How do you install it? How do you build it? How do you do those things? Otherwise, they can reach out to me postpolitical.com is my website. My email is Randall at postpolitical.com. I'm happy to help people with their AI questions. Randall Thompson educating us about AI and scaring the hell out of us about AI. Is that pretty good? It's pretty good. Hey, thank you very much Randall Thompson from postcommunications talking about AI and how it's shifting us these

7:16tools and these these what they call agents are able to do work for us while we're sleeping an incredible fast pace and low cost unbelievably incredible and unbelievably scary at the same time. We need to make sure that our our society is prepared and educated and our kids learn about this so that we can find better ways of making sure that everybody has a job because this is going to displace a lot of people from their work. You're listening to Rocky Roshkowski sitting in for Paul W Smith and WGR. You're listening also to focus.