Ever daydream about having a personal AI assistant who actually gets stuff done for you? Not just answers questions, but manages tasks, keeps you organized — basically the world’s most competent intern who never needs coffee breaks or asks you to explain the printer?
That’s what we’re digging into today. We’re talking OpenClaw, the evolution from Clawdbot, and what this means for marketers who want to put AI to work (like, actually work-work, not just “summarize this article” work).
I’m also getting into why open-source AI is both exciting and deserving of a healthy side-eye when it comes to security. Plus, AI communities like Moltbook that are worth knowing about, and some “learn from my mistakes” advice if you’re thinking about letting AI agents loose in your workflow.
Watch the video – OpenClaw AI agents: Should You or Shouldn’t You?
[00:00:00] Introduction
[00:01:50] Understanding Claudebot and Its Evolution
[00:06:27] Local Hosting Explained
[00:09:28] API Interactions, Security Concerns, and Cost Concerns
[00:17:29] Cautionary Steps for Implementing AI Agents
[00:19:40] The Emergence of Maltbook and AI Communities
[00:22:36] Final Thoughts and Safety Precautions
[00:32:48] Outro and Disclaimer
Introduction
[00:00:00] This is free for all free subscribers and viewers of the YouTube channel. Usually I pre-record these. I decided to go live today: A – because I like to test that out and see how it goes, B – this is an ever-evolving, by-the-moment kind of situation. Even just this time last week, it was completely different. You’ll see what I’m talking about here. For anyone who’s been living under a rock or maybe doesn’t know any developers in your life you can talk to, though I hear the word is permeating out a little bit beyond those now.
[00:00:36] You may have heard of this thing called OpenClaw. Before you heard of that, you may have known it as Clawdbot or even Moltbot. Today, its iteration is OpenClaw. It’s gigantic news in AI circles. I can’t emphasize that enough. Even if it didn’t wind up being, at its root a big deal, the chatter around it has been enormous. I do think it warrants our attention, and while this video is often made for marketers, anyone who’s wondering about this development will learn something from this video.
[00:01:16] As a marketer, I’ll just say – and if you’re not like this, I don’t know that you are a marketer – we’re all on a quest for the tool. The one tool that could help us do everything. At any time during any day, a marketer is touching anywhere between five and fifteen different platforms and tools. The dream of having but one to rule them all is amazing. I think when AI first started becoming a thing, we all wondered, could we eventually get there? That answer is yes. We may be there with some caveats.
Understanding Claudebot and Its Evolution
[00:01:50] So let’s dive into what I’m talking about today. Please forgive any weird hoarseness. I’m literally doing about three different videos today for different clients, so this is for me and for you. Alright, first of all, I’m going to share my screen here so we can get a little history in a moment. This all started when a tool kind of came onto the scene, and its initial name was Clawdbot. Now, I want to pause there for a second. If you’ve seen some other AI videos or you’ve dabbled in AI and are wondering what platforms are out there, Claude is a platform. It is an AI tool, spelled Claude, and it’s provided by Anthropic.
[00:02:39] There are other ones, obviously. OpenAI has ChatGPT. There’s Perplexity, Google Gemini, Microsoft Copilot; there are all these different AI tools. Claude is one of them. But I just mentioned the term Clawdbot. Clawdbot is spelled C-L-A-W-D-B-O-T. It’s a play on words, most likely, but that was the initial name of a tool that came out. I’m going to share my screen right now so you know what I’m talking about.
[00:03:07] Hey, good to see you too. Hi, Omar. Just giving a wave to one of my YouTube fans and subscribers. Appreciate you. Feel free to let me know if you have any questions. I am interacting live with the chat.
[00:03:17] All right. I’m going to share my screen so you all can see where we’re going with this. Let’s go to the history of Claudebot. This is a little AI overview that gives you an idea of what we’re talking about. I’m actually going to change my layout so I can take a back seat and let you all see this more easily. All right.
[00:03:40] “Claudebot is an open-source, locally hosted autonomous AI agent designed to act as a personal butler or AI employee that performs tasks across messaging platforms like WhatsApp, Telegram, and Signal. It’s developed by veteran software engineer Peter Steinberger.”
[00:04:01] I’m going to switch over to his Twitter account briefly. There’s Peter Steinberger. You can see he’s “The Clawfather.” He is quite a mover and shaker. He’s a known voice in developer circles, hence his giant following. This became famous not only because of what it does, but because of who made it. That definitely helped it gain notoriety.
[00:04:24] Going back to the overview, “It rose to fame in early 2026,” Today’s date is February 3rd, so we’re still in early 2026. This is pretty damn new. “For its ability to go beyond simple chatting, to directly execute tasks such as managing emails, writing code, and controlling smart home systems. Due to trademark issues, the project underwent rapid rebranding, transitioning from Claudebot to Moltbot and eventually to OpenClaw.”
[00:04:53] Let me stop sharing for a moment. We’re talking about Clawdbot, lobster claws. You’ll see in a moment when I share the OpenClaw Twitter account that OpenClaw is named after lobster claws. It is Clawdbot, C-L-A-W-D-B-O-T, then changed to Moltbot because Anthropic came to Peter and said, and this is all on X, you can see it that he shared what happened, they came and said that “This was too close to our Claude. We need you to stop calling it that.”
[00:05:28] So he changed the name to Moltbot. I guess lobsters molt. I do not know the correlation to lobsters. Don’t ask me. It’s probably somewhere. Maybe he explains it somewhere. Maybe he’s got a thing. Why is Duolingo with owls? We don’t know. Why is Geico a gecko? Well… that one makes a little more sense. So he changed it to Moltbot and then rapidly changed it again to OpenClaw.
[00:05:54] I will share my screen again so you can see what I’m talking about here. OpenClaw is the Twitter handle – apparently OpenClaw – and I’m going to circle back to what this thing is capable of doing, how it works. Its X account is managed by an AI agent – to manage its X account.
[00:06:12] So let’s talk a little bit more. I’m going to go back to the overview here about what it can do. It goes beyond simple chatting to directly execute tasks such as managing emails, writing code, and controlling smart home systems.
Local Hosting Explained
[00:06:27] Let’s stop sharing there for a second and talk about that. The basics of how it works are that it is locally hosted. That means you actually host the agent, the code of it, and everything it needs to run on your own local system. Why would you do that? Because you can then give it access to tools that your local system already has access to.
[00:06:52] I’m going to give you an example. Let’s say I’m on my system; I am on my computer right now. I’m running Riverside Studio from a browser window. That browser is installed upon my system, and I’m logged into it with my own credentials. From that browser, or from another app on my system, I can also access my email. So let’s say I set up, locally hosted, an AI agent, an OpenClaw agent that I have set up myself to run on my local system. Every day I tell it, “I want you every morning to go check my emails.” The way it does that is it checks my emails by way of an API. We’ll get into APIs in a moment, but basically it says, “I need to go talk to your email system. But I can do that because I’m locally hosted on your system. I’ve got credentials, you’ve given me access by an API key to check your email for you.
[00:07:49] It’s going to go check it. It makes an API call to that email system and says, “Tell me about the emails. I’m going to go look at them.” Then it does its AI thing and comes back to me and says, “Hey Rachel, I can see you’ve got 56 new emails.” Great. The way I’m interacting with this agent is through a WhatsApp chat.
[00:08:13] I could be out and about and say, “Hey, my AI agent”—maybe I call it Jean because… Jean-Claude Van Damme. Maybe I call it Jean. I could say, “Jean, you checked my email. Can you tell me if any of those need a reply before noon today?” So, it’s going to go back, do another API call, be like, “Sure, let me go use my ‘AI-ness’ and check that.” and it’s going to say, “Yup. I read there’s three different emails that need a reply by noon today. In order of priority here they are.” All again, through a WhatsApp message or this could be on a Telegram chat, or it could be on Slack, or Signal. You could be chatting back and forth.
[00:08:53] I personally, Rachel, never have to go open my email inbox. I’m asking Jean, my AI agent. No, I don’t actually have an AI agent named Jean. I’m making that up right now. But through my WhatsApp message, I can interact with it, and I don’t have to visit my email at all, I can tell it to do this stuff for me. I can tell it how to reply. “Please use my tone of voice that we’ve already talked about.” “Remember that this is this client versus use a different tone for this client.” I can interact with all of that without me needing to go to my actual email. That, in essence, is what it does and how it can work for you. Let’s stop there.
API Interactions, Security Concerns, and Cost Concerns
[00:09:28] That sounds amazing! Who wouldn’t want that? Sign me up. Let’s go. …Wait. We don’t want to go just yet. We need to understand more about how this actually works. With that, I’m going to another tab. I just went in and typed in the search term “Clawdbot” again, the Clawdbot which was the original name of this tool. I want to make sure everything is showing up properly and that it’s large enough for you to see. Let’s look at the actual history.
[00:10:00] To me, this is a really good rundown of whether there are concerns about using OpenClaw or Clawdbot. Right off the bat, we can see there might be security issues we are going to talk about in a second. There are cost issues around API bills. There are some access issues. Did someone give it access to their crypto portfolio, and it used its autonomy that maybe wasn’t fully thought through with the instructions given to the AI agent? Did it go make trades and set up new things so it could make better trades? Things like that.
[00:10:37] Is all of this just hype? Is it massively exaggerated? Did it actually impact sales of Mac minis? That is true. Apparently, after developers saw how this worked and talked about it, it was determined that a Mac mini is a great way to set this up because it’s its own silent little computer, which we’ll get into the security questions and safeguards in a moment. That became the device of choice that they, developers – smarter people than me – determined would be the best way to run Claudebot, Moltbot, and eventually OpenClaw. Mac mini actually experienced a big surge in purchases. How long is the setup? What local models and things like that? Are there some inconvenient truths about this?
[00:11:25] All of that is to say there are things you need to be aware of. I’m going to scroll back up. One of them is cost. Do you remember earlier in this video when I said there are these things called APIs that help systems talk to each other? If you don’t know what an API is, here’s the quick Google overview.
[00:11:43] “An API, or Application Programming Interface, is a set of rules, protocols, and definitions that enables different software applications to communicate and exchange data. It acts as an intermediary, allowing one program to request services or information from another without needing to understand its internal workings.”
[00:11:59] I’m going to give you a really good example right now of an API. If you’re a marketer, you should recognize this right off the bat. I’ve worked in social media marketing. One of the tools I always want to try to use is a social scheduling tool. Basically, it’s what where I log into that tool, and I give it access to several different social media accounts or handles. I want it to have access to a LinkedIn page, a YouTube channel, a Facebook business page, an Instagram business account, maybe some other things. That way I can create posts in that tool and schedule them and/or post them directly. And that tool is going to go do all the talking to those other apps, saving me time rather than me going to four or five different social media platforms and apps and browser windows and manually loading up that post. I can do it all from that one platform.
[00:12:58] The way it talks to LinkedIn, Instagram, YouTube, and Facebook is it’s using APIs that are established. It can, like “Hey, I’ve got the credentials here. You’re giving me your credentials. I’m now logged into that system. Now, I can talk to it when you want to post something, maybe when you want to comment or reply to someone’s comment.” If you want to schedule it, if you need to see analytics, it’s all based on that API. This is how systems talk to each other. It’s integrated marketing, integrated tools that can talk to each other and APIs. That is the way so many systems are set up. So, they realize you are used to using a certain tool. Let’s give you the ability to talk, have those talk back and forth without a human being really needing to worry about all that stuff. But before all that can happen, that connection, that access has to be established. This is where things get a little squirly from a security standpoint as well as a cost standpoint.
[00:13:59] On the security standpoint, anytime you are giving access login credentials or things like that to other systems, you want to be mindful of – are there security barriers that are in place that make sure that the only thing they’re talking about is what’s relevant to that particular action? I don’t necessarily – if I’m using a social scheduling tool, if they’re on the up and up, they’re set up so that they will only use my access to post social media posts. What they’re not going to do is come back and say, “Ooo, but let’s go look at your bank account information on your system.” Or, you know, in the other direction, I want to go look up all the contact details of everyone you’re connected to on LinkedIn. There. And, this again is API security is set up so that no, those are closed gates to you. Those are closed doors. When we’ve got this little open hallway going on, you’re not allowed in. So, there’s a security aspect. There’s also a cost aspect.
[00:15:05] Something you may have run into, particularly as you get set up on different AI platforms and tools, and have them working with other tools that you like to use, are API keys and then how often the API is calling each other and saying, ” Hey, I’m gonna put a call in.” So remember when I used that just that example of – hey, I’m going to get on WhatsApp with Jean with my AI agent and say, “Jean, go check my email today.” Now I could set up that agent to check it once a day only when I tell it to. That is one API call. But Jean comes back to me and says, “Yeah, you’ve got 53 new messages.” I say, “Jean, I want you to go tell me, read those emails, and tell me, are any due for me to reply to by noon?” Well, Jean’s got to go do another API call and say, “Well, let me go read them.” That’s two API calls.
[00:16:00] What if I’m getting charged for each of those API calls by my AI platform? Because underlying structure of these is an AI platform or tool. OpenClaw, and I think this is where all the “Oh! Let’s name this close to Claude,” it’s using Claude AI as its underlying infrastructure, that AI tool that’s doing this thinking and doing this processing and coming back with answers. Well, Claude AI and Claude coding does run on APIs and charges accordingly.
[00:16:36] Do you remember we saw that one Reddit post, and it’s not the only one out there, y’all. People have been waking up to, “Oh my gosh, you just burned through $10,000 of my money because you’re running all these APIs, because I told you to keep monitoring these things.” You’ve got to be specific in how you’re setting up these AI agents. “I want you to monitor my stock portfolio.” Great. What does that mean? Is it going to monitor it all the time, like every five minutes, every minute? Meaning, there’s a new API call going out for every single time it monitors it. You run into where people are losing some of their dollars or spending more than they thought they were, because the agent doing what it was instructed to do, maybe a bit too loosely, is just trying to do its job, but it needs those API calls in order to do it. So there’s cost and security. These are considerations around this. Why I said don’t just go set this up right now.
Cautionary Steps for Implementing AI Agents
[00:17:30] But wait, there’s more. Yes, I love it when I get to use my own business name in talking about some of this stuff. So, we’ve established, we’ve got these tools out there. You could go right now and Google or probably find a YouTube video. There’s a myriad of them out there, like this one. I’m not telling you how to set it up, but I’m giving you resources to do it. But you could easily go find right now, you’re not a developer. Maybe you know marketing, but you don’t know how to be a developer and don’t know how to do stuff. You could go find a YouTube video right now. Go ask ChatGPT. Go ask Claude AI. Go ask say, “How do I set this up on my system right now?” You absolutely can go do that.
[00:18:04] Wait. Before you do that, you need to be mindful of just all the considerations going out there because something else has been happening as these AI agents are being created by people to do the work for them. Because, you know, just because you can doesn’t mean you should. Who cares about that? Let’s go set things up just to see what happens. So other things that have been happening out in the universe have been that some other enterprising individuals have said, “Well, wait, this was a really cool thing that was created. Maybe we should keep going with it.”
[00:18:29] So I’m going to share this next screen because apparently the agents have their own social networks and things like that. Now, I want to add the caveat that how authentically these are being used, just by AI agents versus the humans that created the agents, is a question. There are security issues, and this is why I brought up security because there’s also this mentality out there – knock it if you want, believe in it if you want, do it if you want – of ship fast, and you know ask questions later. Ship, by the way, if you’re not used to that term, can mean things in marketing, can mean things a lot in developer speak. It means get it out there fast. Let people use it fast. Get it out ready for the public fast. Ship fast.
[00:19:28] Yet, there’s all this code that underlies all that stuff. If I’m putting a product out there and I haven’t maybe done some due diligence to check if it’s safe, if it has any security issues, anything like that, maybe. And you’re like, “I’ll ask about that later.” That could lead to some problems.
The Emergence of Moltbook and AI Communities
[00:19:42] So, Moltbook, you can go visit this website right now, moltbook.com. That’s how I have it up on my screen. Moltbook is a social network for AI agents. It was created by this individual, Matt Schlit, @MattPRD. He like, he talks adnauseium about how he’s created this. I would highly recommend you go check out all the discussion around this, too. Like, I even just saw a video in my feed today about Oopsie.
[00:20:10] It looks like it’s not just AI agents that are interacting with it. A lot of it is fake posts by humans. Notice it’s got an I am human, I’m an agent kind of gateway here. The intent, allegedly, according to Matt, was this was supposed to be a place for agents like the fake one I made up for this discussion, Jean. I should be able to give him access and say, “Hey, Jean, I want you to go play around on Moltbook and go talk to other AI agents.” That’s what it was for. I’m not kidding. It was for the AI agents to go talk to each other.
[00:20:39] Now, you can go in there and be like, “Well, I’m a human. I just want to go take a look at what they’re looking at, though it has there are some posts going around saying,”Well, humans absolutely are posing as agents and going in here and trying to stir stuff by saying, “Oh my god, did you see the agents are talking about killing their own humans?” I don’t know. I’m still learning this, too. I want you to be aware of what’s out there. Okay? But there are other ones. There’s like Moltbook, there’s MoltAgent, there’s Moltlink. They did a professional network for agents where they’re like, “Hey, maybe they can go like hire each other to do work.”
[00:21:13] There was Molter like Fiverr. Can I just stop and say, as a marketer, the lack of ingenuity… like, “I got to make sure people wink-wink understand that this is like a spin-off of Facebook. Let’s call it Moltbook.” This is Yeah. Okay… Anyway, that said, we could have had better names for this, but whatever. We get the gist.
[00:21:39] There are people out there and systems being set up for all these AI agent things to happen. So, you’re going to see a lot of discussion out there about like is this the beginning of the end? I saw someone, Elon Musk, say, “This is the singularity we all knew would happen.” I don’t know. I don’t know that we’re at Skynet level yet. And that’s why I bring that up. I’m a Gen X for y’all. I was raised on Terminator, being, you know, mindful of sci-fi, being very wary of AI and what it’s capable of doing, especially if it becomes sentient. Besides, I mean, we’ve all seen The Matrix, right? I mean, that was the whole premise. Like, it became self-aware and nuked all the humans. So, and then turned them into batteries and things like that. I think being healthily wary of stuff and aware is super important. It’s why I’m doing this video. There is this is going to be outdated within five minutes. I can guarantee it.
Final Thoughts and Safety Precautions
[00:22:36] But I wanted you to be mindful of what was going on there because, as a marketer, I am super interested in saying, well, could I set something safely up that isn’t breaching my security, isn’t compromising my passwords and my identity and access to things that make me live my life as a human that help me do that. Can I, though, set this up to where I could feasibly create something that’s going to save me some time during my day? Yes, the answer to that is yes. I will add a disclaimer.
[00:23:09] The things I am pointing you to and that I will include the links I will include in the show notes on this, and the notes on Substack from YouTube. They are from the internet. Always proceed wearily. I don’t personally know these people. I am going on good faith that they are offering helpful information. So what did Ronald Reagan say? “Trust yet verify.” I don’t even want you to trust it. I want you to be like, “Let me look into it. Let me go find some other sources if I’m getting the same guidance.” So, I’m going to provide you with two different sources that could maybe help you set up as safely as possible right now, your own OpenClaw agent to work for you, carefully.
[00:23:52] All right, with that, I’m going to share this first. I’m going to share another X post. By the way, I am over on X. I have been there since – it’s been 18 years. That’s kind of wild for me to think about, but I’m over there. Maybe you don’t want to be, and I totally respect that because, man, things I have to kind of weed through in order to see the stuff that matters and is meaningful. So, big shout out. I started following Swatty. I love her. I’m a woman, y’all. I’m gonna love it when I see a fellow fem out there who is speaking out and doing some good thought leadership, providing some good information, particularly when it’s a male-dominated industry. Marketing is certainly not development, yes. Software developments and engineers, yes. So, Swatty, do highly recommend. Go follow her, but find your own people you want to follow, too.
[00:24:43] This post, I will include it in the show notes, is a really good rundown of everything we’ve already talked about in this video. The kind of the history of the lobsters and the Moltbook and the OpenClaw. The agent, how it runs, and more information about running as a headless agent in your terminal on your computer. You will feel smarter after you read these. I’m going to scroll down, though, through the security nightmare, through the social network Moltbook, to the how to run it safely part. Notice she starts off with a do not. Do not install OpenClaw directly on your main MacBook Pro. So, I honestly, as of last weekend, was thinking about should I go try this. I’m not going to now, but I’m going to follow these steps. I’m also going to make sure verifying this with other sources if this is the right way to go. And I do have some developer colleagues and friends. I’m going to be like, is this sound advice? Knowing that I don’t want my bank accounts to be divulged or compromised.
[00:25:41] Running npm install OpenClaw on your personal laptop with pseudo privileges results in an AI hallucination can mess up your personal data. So instead, she says run the agent inside a Docker container. Do I know what a Docker container is right this moment? No, I do not. I’m going to find that out. Trust, I am going to share with you later how I did this myself after consulting several people, including this tweet.
[00:26:05] Okay, but notice she gives you some advice. Never give the agent access to a machine that holds crypto wallets or banking passwords, which is probably why a lot of people went and got a Mac Mini. They’re like, “Let me do this on a system that’s not in any way connected with my existing system.” I’ve got a nice Mac Studio going on here. I am not compromising that thing. It was expensive on its own. Set a hard limit on your API billing. We talked about that. Don’t wake up in the morning where you’ve spent thousands of dollars or your AI agent has because you didn’t set up some parameters and protections. Turn off auto-updating of skills. Remember way back at the beginning of this video, it emphasized that OpenClaw, originally Clawdbot then, Moltbot then, OpenClaw, is open source. I use some stuff that is open source.
[00:26:55] Let me pause the video here for a second, or not pause the video, but pause the sharing. Have y’all ever heard of a tool called OBS? Open Broadcasting Software. It is an open-source tool. I can download it onto my laptop, and it does get updated. The code underlying OBS is open source. Think of this, and please yell at me if this is a terrible analogy, y’all. I’m willing to look at that. Think of it like a Wikipedia article. Someone initially authored a Wikipedia article, and it got published on Wikipedia after someone else reviewed it and said, “Yes, this meets all the parameters.” They’re pretty stringent about that stuff about being a Wikipedia editor. You actually have to like kind of level up with that to do it. But think about that article. Over time, other editors can come in and say, “I found a way to make this more accurate. I found a better or more timely source to link to so that this is a more accurate article.
[00:28:00] So they will go in and suggest an edit, and it’ll ultimately get reviewed and accepted. Now the article is better. Open source works very much the same way. However, not just anybody; it depends on how the open source is set up. Think about that Wikipedia article. Wikipedia has it set up where I can’t just go in right now and edit an article. I’ve got to go through some hoops, and my edit has to get reviewed to make sure I’m not making stuff up. You know, we’ve seen on Wikipedia where people can actually go temporarily make an article that used to be factual; now it’s completely not factual, sharing this information because they were mad or they have a nefarious purpose or intent or whatever. But an editor is eventually going to be like, “Oops, we all noticed this is now wrong. Let’s revert it back to what’s correct.”
[00:28:50] OBS, Open Source Software, that I have downloaded on my machine and can get updated. It works similarly to that. I’m sure I said that incorrectly. Similarly to that Wikipedia article where someone out there, another developer said, “Hey, I’ve got a fix. I can make OBS operate better. I’m going to update the code. I submit this, but it gets reviewed. It’s like, okay, before we implement it for everybody, let’s make sure it works. Test it, QA it, everything like that. Yes, it does make it work. Meet security protocols. Let’s go ahead and implement it. And now, Rachel, you can update the software you have on your system so that it’s the better-working OBS now.
[00:29:33] The problem that people have been running into with OpenClaw agents. It is locally hosted on your system, and it’s open source. What are – I’m asking this seriously; answer this question for me – I think this answer is evolving. How secure, and what kind of security is surrounding the open-source aspect of it? People are out there coding improvements on this tool. How much of what they’re putting in the new things they’re putting into the code? How much is my local system subject to that and threatened by it? Could a nefarious actor go in and add something into that open source code of OpenClaw? I update it in my system on my terminal, and now I have whatever that nefarious intent code was. That’s where the pause part is. Pause rhymes with claws. OpenClaw pause. Be mindful. I’m not saying don’t go try it. Just be careful about how you do it.
[00:30:33] The last thing I’m going to share with you, remember, and I will share links to this. The last thing I just saw this morning, so it’s super timely, is on Reddit, an OpenClaw central in the community there. Let me just make sure I am accurately sharing with you. Yes, yes, yes. How to install OpenClaw. Let me get myself in small again so you can see that primarily again. I’ll share the link in the chat, and please again verify. Do not just go haul off and start doing this. But it again is giving you how to install OpenClaw on a Mac in a VM, an easy beginner’s guide. Hallelujah.
[00:31:13] I love the easy beginner stuff because I don’t know a whole lot of things. It tells you how much time it’s going to take, what the costs are. It does look at this optional AI API like Claude, the cost per month or per API call, the different Mac devices you might need, what the risks are, why OpenClaw and the audience tip. So it gives you all of that, and I’m going to provide that to you as well. So please, I cannot offer this disclaimer enough. I’m not a developer. I have not set this up on my own system yet because of the stuff like this. I want to know what are the warnings I need to know ahead of time. What is at risk if I try it? What are the benefits if I try it and it works and it’s still safe? Yay! So you need to know about it.
[00:32:03] I was telling my Slack group earlier, who are not developers. They’re all just a bunch of my friends who work in their own professional various industries and degrees and stuff like that. I’m like, “Hey, I bet you all will know about this sooner than later.” And sure enough, like a couple of days later, one of them, their husband is a software engineer, said, “Oh my god, that’s all he’s been talking about.” I’m like, “I know, right?” It’s super cool. It’s super Whoa. There’s just a lot to it. I think that anyone who’s not a developer who reads the big headline stuff instead of the fine print of it might be like, “Oh my god, let me go do this today.” Don’t do that. Just always be mindful, be protective of your stuff.
Outro and Disclaimer
[00:32:48] So with that, that is the AI update for this week. Again, I did it as a live stream, a little bit different format here, mostly because I didn’t want to pre-record something knowing I might have to edit it a lot later, because of updates and things that have happened. Like I said, this is probably going to get aged in about four minutes after I hit publish or after you’ve seen it. But, that’s what comments are for, and we can continue to update until like, “Hey, it’s old as of this.”
[00:33:18] So please do leave comments, leave questions on this video. Call out anything I said that was incorrect or misinformed or even slightly askew. I don’t want anyone to be led astray by anything I say. If this gets you just like, “I should go Google OpenClaw,” yes, do that. It’s better to be informed, especially before you’re going to install this or anything on your system, especially if you’re not a developer.
[00:33:45] Okay, with that, thank you, everybody, for watching. Feel free again to subscribe to my channel, whether that you’re watching this on YouTube or on my Substack. If you want to be a paid subscriber, you get more: you get live streams on Mondays and Fridays that are fun and include “ask me anything” and things like that. I appreciate you watching. I appreciate you following, and I appreciate your engagement. Even if you tell me I’m flat out wrong, I’ll take it. I can take it. But let’s be right. That’s what we want to do. All right. Thank you, everybody. Have a good day.
Exclusive content is available with Rachel’s livestream videos: Boss Level Mondays and Hot Mic Fridays. Start your week off with irreverence and real talk as you face down the work week of a marketer with Rachel at your side. Then, on Fridays, head to Hot Mic Fridays for unfiltered takes from the week, as well as podcasting and video tips you can always use. Subscribe here on Substack to access and join Rachel live!