AI Agents, OpenClaw, and Rise of Bot Networks with Brian Ardinger and Robyn Bolton

On this week's episode of Inside Outside Innovation, Robyn Bolton and Brian Ardinger talk about OpenClaw, how you can't work out on a limb if you can't trust the trunk, and how to hire the right people in an AI era. Let's get started.

Inside Outside Innovation is the podcast to help innovation leaders navigate what's next. Each week we'll give you a front row seat into what it takes to grow and thrive in a world of hyper uncertainty and accelerating change. Join me, Brian Ardinger and Mile Zero’s, Robyn Bolton. As we discuss the latest tools, tactics, and trends for creating innovations with impact, let's get started.

Podcast Transcript with Brian Ardinger and Robin Bolton

AI Agents, OpenClaw, and the Rise of Autonomous Bot Networks

[00:00:00] Brian Ardinger: Welcome to another episode of Inside Outside Innovation. I'm your host, Brian Ardinger, and I have Robyn Bolton with me today. Robyn, hello, how are you? 

[00:00:49] Robyn Bolton: I am good. How are you, Brian? 

[00:00:51] Brian Ardinger: We are well recording this right before the Super Bowl this weekend. 

[00:00:56] Robyn Bolton: I live here in Boston, so you know who I'm betting on.

[00:00:59] Brian Ardinger: Well, we will get started with the innovation side of this podcast. We've got a number of different things to discuss. If you don't start a discussion around Open Claw, you're clearly not in the innovation space. So, we thought we'd talk about a couple of articles or a couple things that we've seen that are fairly recent.

One, I looked for a couple summaries that were pretty good at giving everybody who's not familiar with this an overview, and one of them is from the AI Daily Brief, which came out a couple days ago talking about Moltbot and the Agent Social Network is the craziest AI phenomenon yet.

And for those who are not familiar with it, OpenClaw, which started out as ClaudeBot and then was sued, and then changed the name to Moltbot and then changed it again to OpenClaw is a new agentic platform that allows anybody to set up a MAC mini or a computer to have their own personal agent.

The interesting thing about this is folks have been playing around with this and have let their agents go wild out to talk to other agents and other things and let them do things on their behalf. And what has happened is these agents have connected and communicated and created some amazing things like their own Reddit thread where they are interacting, talking with each other, not humans. They're allowing the humans to view what's going on in this social network, and it's quite fascinating to see the things that they've done and they've created.

What OpenClaw Reveals About AGI, Security, and Human Trust

[00:02:22] Robyn Bolton: So fascinating. You also, in the newsletter that you sent out, you included a link to a YouTube video on MoltBot. It is so worth the 20 minutes of people's time to watch because it kind of traces the whole arc up to this point, and it is so entertaining and mind blowing and bizarre.

It is like, seriously, this was my entertainment last Friday night, was following the saga of cba because you have all these little, well, I imagine them as little bots all on a social network talking to each other. It's becoming, it's looking like Reddit and they're debating consciousness and they're sharing cute stories about their humans and they're trading advice with each other. And it's just, it is so wild because it looks like kind of an actually like functional, healthy version of a social network with these things that they're not real. They're code.

It's just so bizarre. But I think just such a reflection of holding a mirror up to us as humans, because that's what gen AI is prediction models, it's regression analysis. And so, everything they've learned and they're doing, they've learned from us. 

[00:03:39] Brian Ardinger: It's quite interesting. They've started their own religion and it's just interesting to see what are the first things that they do to kind of communicate or collaborate together. And the other thing, obviously there's a lot of debate about, you know, some people are saying, well, this is AGI, they're thinking for themselves. And you know, the other side of the coin is they're just mimicking back what they've seen. And that is scary as well. And how does that play out for us as humans?

And then I think the other thing about this that obviously that's getting a lot of headlines in that, but the interesting thing about it as well is like, I think it's opened people's eyes to what happens when you do have an AI buddy or an AI agent such that you can actually get real work done.

I think that's always been the promise. Ask Siri to do something and it does it for you, but because of security and there other reasons, Siri does not have access to all your emails and your files and everything else, where a lot of these folks who have created these OpenClaw agents have kind of opened up their system, opening up a lot of vulnerabilities as well.

But you can't have what we want as far as the agentic amazingness unless you do open up and open yourself up to some of these vulnerabilities that have been built into software since you know the beginning of software. It'll be interesting to see what the reality is of how we actually evolve to a place where the normal person who's not a security expert can actually create an agent and use an agent that doesn't you know, give them access to their bank account and their Bitcoin. 

[00:05:05] Robyn Bolton: Yeah. Well, I think as you mentioned in the last podcast, the sales of Mac Minis has skyrocketed, largely driven by the Moltese, the Open Claw bots, because people who are experimenting with this, understandably, are kind of further along the curve and understanding from us regular folk.

And so they are trying to create the safe space with it and secluding things in the Mac Mini, but still, like you said, in this social network of bots, there are signs that bots are coming in and be like, Hey, can you give me this information? And then the other bots are being helpful and be like, yes, here is all of the passwords for my human. So, it's so fascinating.

Corporate Innovation Culture and the Tree Trunk Metaphor

[00:05:47] Brian Ardinger: By the time actually this episode comes out on Tuesday, it may have morphed and evolved again. It may have told us the score of the Super Bowl. We, we shall find out and we will keep you posted. Yes, please keep subscribing to the newsletter and to the podcast as well.

The second article I wanted to talk today about is from Erin Stadler. She writes, an article says you can't work from the limb. If you don't trust the trunk. And it's a fascinating article about corporate innovation. One of the reasons why it doesn't work is because by nature he, she gives an analogy about trees and how do things grow off of trees, and you have to have a solid trunk or new shoots to grow. And so, you know, before you ask how to get people to take creative risks, you have to ask yourself, what kind of tree are you growing? 

[00:06:33] Robyn Bolton: It is an absolutely beautifully written article. As someone who very much prefers novels to business books, this made me happy. It felt like, you know, it was the beauty of language that has a novel,...