THIS WEEK IN AI
Is AI a Bubble That Will Burst? Our Hot Take.
With all the chatter around AI, people are asking, "Is AI headed for a crash like the dot-com bubble?" When they asked, I had to stop and think, and then do a little research, because the dot-com era started in 1995 and crashed 2000, and one of us was barely 8 at the time.
Let's explore and compare then and now.
Remember Pets.com? (No worries if you don't, we have the receipts.) It was an online pet supply company that launched in the late 90s with a massive marketing budget, a Super Bowl ad, and a sock puppet mascot that was iconic, and maybe a little weird. Here is a snapshot of their website from Nov 28th 1999, thanks to the Wayback Machine.
They were spending more money to ship a bag of dog food than they were charging customers for it, and they had no real plan to fix that. Within a year of going public, they were bankrupt.
And that wasn't unusual. The dot-com era was full of companies like this: People assumed that if you were a company with a website, it was enough to become a million-dollar company. Investors were pouring money into ideas that didn't have the infrastructure, the logistics, or a real path to profitability behind them. When reality caught up, it all collapsed AKA the dot-com bubble burst.
So is AI the same thing?
We don't think so, and here's the clearest way we can explain why. The dot-com bubble was built on a technology that was real but wasn't ready yet. Reliable broadband wasn't widespread. Online payments weren't trusted. The foundation wasn't there to support what people were trying to build on top of it. Or with the wider reach companies had, they couldn't keep up with demand, or get the profit margin there....and... POP.
AI is different because the tools are already working. Last week we created a tool to write your website for you. That's not a future possibility. It actually happened. Businesses are already using AI to write content, handle customer questions and calls, build proposals, and automate work that used to eat entire afternoons. You don't have to imagine a world where this is useful because we're already living in it.
Think about when spreadsheets replaced paper ledgers. Nobody called Excel a bubble. It just made existing work faster and more accurate, and businesses that adopted it early had a real advantage over the ones that waited. AI is doing the same thing, just at a much bigger scale and a much faster pace.
Will some AI companies fail? Absolutely. There is always the possibility that someone invests in a company that isn't well thought out, run, and tanks. Will there be consolidation? It's already happening. But the technology itself is already woven into how real work gets done. That's a very different situation than a sock puppet selling dog food at a loss.
OPINION
What If AI Created Jobs Instead of Taking Them?
I'm sure you have thought about or heard a friend make the comment, AI is going to take all our jobs. And maybe that has gotten into your head, and you've even felt a little guilty about it. Wondering, if you start using AI to help write your content or handle something that used to take you hours, will you be taking someone's job?
So glad you are a real human with a heart for asking the question. It's worth actually sitting with that question instead of brushing it off.
We could think of it that way, or we could think of it through a different lens. As AI gets better at specific tasks, we won't always need a human to do that exact task anymore. That part is true, and we're not going to pretend it isn't. But what's also true is that one person can now oversee what used to require ten. In the development world, instead of needing ten developers to manage ten projects, you could have one developer overseeing all ten because AI is doing the heavy lifting on execution. Those ten developers don't disappear. They each get more done. More gets built in less time.
Tony Robbins made this point recently in a conversation about AI, and it stuck with us: someone has to manage the AI. The skills don't become worthless. They become more powerful because the person who has them can now do so much more with them.
And that's where it gets exciting for small business owners.
People who have had ideas sitting in their heads for years, ideas that felt too big, too expensive, or too technically complicated to actually build, are now building them. Without having to hire a developer. Without knowing how to write a single line of code. Just a clear vision, a willingness to learn the tools, and a lot (I mean A LOT) of conversations with AI to bring it to life.
That's going to create a wave of new entrepreneurs and solo business owners that we've never seen before. And yes, it will get crowded. When the barrier to building something drops this dramatically, a lot of people are going to try. It's why people say "no one is successful " when it comes to direct sales. Well, when it costs $150 to start a business, everyone does it. And when they quickly realize there is real work involved to get it to grow, they bail. The same will be true here. At $20 a month to build an app, there will be a lot flooding the market.
The ones who thrive won't necessarily be the ones who moved the fastest. *phew, you're not behind* They'll be the ones with strong branding, a real marketing strategy, and the consistency to execute it over time. (Reality check) The idea alone won't be enough. It never has been.
Which is, not coincidentally, exactly why you're here. Because that part, the brand, the strategy, the marketing, that's what we teach. It's what we build for our clients. And it's what we're showing our community how to do, with AI making the whole process faster and more accessible than it's ever been.
The opportunity is real. The question is just whether you're building something worth finding.
HOW TO USE AI THE RIGHT WAY
You Don't Have To Master AI Today. Start With The Language.
I was sitting at my kitchen table while my husband's friend (who's been building AI Agents for the last 2.5 years) was deep in a conversation about AI. He was saying terms and abbreviations like everyone in the room should just know what they meant, and things were going completely over my head. I couldn't visualize what he was describing. I couldn't connect it to anything in my life or my business. It just floated past me like I was in a foreign country listening to a language I had never learned.
I started typing the terms into my notes app as he talked so I could look them up later.
Here's the thing about hearing a language you don't speak. You can't do anything with it. You can't participate in the conversation. You can't have ideas about what's being said. You can't connect the dots because you genuinely don't know what's happening. It's not that you're not smart enough. It's that the words don't mean anything yet.
So I looked them up. I didn't memorize everything. I didn't get a full technical understanding of every term. I just got enough of a concept to have a reference point. The next day I searched for and listened to an AI podcast. When they mentioned a few of those same terms without explaining them, I actually knew what they were talking about! (I felt so smart!) MCP, for example. Before I'd looked it up, it was just letters. After, I understood it was describing how one platform connects to and communicates with another. That one small thing made a whole section of the conversation suddenly make sense.
And once I could follow the conversation, I could actually think about it.
Ideas started forming. I could hear someone describe a workflow and start wondering how something like that could work for our clients. For a while before that, I had this vague sense that AI could probably help our business somehow, but I couldn't get more specific than that. I didn't know what I didn't know, so I couldn't even figure out where to start.
The more I started understanding what was being said, the more I started seeing how much we could actually do. How we could use AI on the backend to help people become more efficient. How we could put everything we know into AI tools so our clients could use them to create what they need, without needing us to do it for them every time.
And that last part is the thing that blew my mind a little.
The idea of being able to give you a done-for-you experience to build your website and marketing material at a DIY price. That's what we're working toward. And it started with sitting at a kitchen table, pulling out my notes app, and deciding to look up the words I didn't understand.
Most business owners aren't using AI because they don't yet have a clear picture of what it can actually do. And you can't generate ideas for something you can't visualize.
You don't need to learn Claude Code today. You just need to start getting familiar enough with the basic terms and concepts that you can follow the conversation. Because once you can follow it, the ideas come on their own.
We wrote a blog to help you do exactly that. Plain language definitions for the terms you're going to keep running into, organized by how they actually come up in real conversations. Readers have already been commenting that it's one of the most helpful things we've published.
Bookmark it. Come back to it when a new term shows up and you want a quick understanding.
And if you want a place to bring your questions, talk through how this applies to your specific business, and brainstorm with us in real time, that's what our group coaching calls are for. Twice a month we get on a live call and have exactly this kind of conversation. What tools we're using, how to apply AI to your marketing, and whatever you bring to the table. It's the most accessible way to get direct time with us outside of a one-on-one strategy call. AI Advantage members get special member pricing at $30 a month. That's $67 less than the regular rate.
If you are an AI Advantage member, the coupon code will be waiting for you at the bottom of this newsletter. If you aren't yet, you can join here for just $5.55 a month, then come back to this newsletter to reveal the coupon code!
THIS WEEK WE USED CLAUDE FOR...
A Running List Of Almost Everything AI Did For Us This Week
This is not exhaustive, but it will give you a real picture of how Claude showed up in our week. The good parts AND the humbling ones.
From Krissy's week:
- Co-wrote two blog posts in about an hour. (Saved 4 hours)
- Wrote a three-part social media series and built it directly into a Canva carousel. (More on this in the bonus section below)
- Wrote a full client proposal, a follow-up email, and a project debrief with next steps, all from one discovery call transcript. (Saved 3-4 hours while making the client experience better. Looked like we spent all day on the presentation. We didn't. *Thanks Claude*)
- Helped rethink a sales page that was too educational and not persuasive enough by shifting the focus to objections instead of features after my instruction. (Gave an outside perspective without us having to get outside help!)
From Claire's week:
- Pulled together a comprehensive, prioritized to-do list for a client's VIP day directly from weeks of onboarding notes and transcripts, so we could walk in on day one already knowing exactly where to start. (Saved 2 hours of re-reading and archaeology through old notes)
- Joined us in our team meeting to help clear an inbox backlog, scanning past context on an active client project so every reply was informed and accurate without us having to manually retrace months of history. (Saved about an hour)
- Wrote every single photo label and alt text for two client websites using a custom skill we built that factors in which page the image lives on and what the surrounding copy says, so every file name is SEO-optimized and every alt text pulls double duty for ADA compliance too. (Took an hour to create the skill. Saved about an hour and made sure nothing got labeled "image-1234" and called it a day)
From our Project Managers week:
- Used Claude Cowork to run a detailed report of all 12 current clients before our meeting to discuss to-dos for each client. The MPC ;) allows Claude to view our Clickup, every note, and email between clients and pull data and information. (Saved 1-2 hours)
- Used Claude to craft emails post-meeting based on our meeting transcript. (Saved 1-2 hours)
And now for the part that didn't go so well, because we promised we'd be honest with you.
- Spent a chunk of time going in circles trying to get Claude to troubleshoot a coding issue inside Kajabi. It kept giving us confident-sounding answers that didn't actually work, and we kept trying them anyway because surely the next one would be the one. It wasn't. (Lost a couple of hours we are not getting back. You're welcome for taking that one for the team.)
- Hallucinated. During one of Claire's client projects, Claude added a section to a website update that was a blend of two different ideas nobody remembered discussing. It sent Claire down a 15-minute rabbit hole trying to figure out where it came from before she went back and rewatched the recording. It just made it up.
AI is not perfect, and it's important not to put that pressure on it. Just like a human can make mistakes, so can AI. It works best when it has full context and room to actually execute. When you're working inside a platform with walls, know when to close the laptop and call support instead. And always verify or double-check. The output is only as good as the information it has, and sometimes it fills in the gaps with something that sounds completely reasonable and is completely wrong.
Estimated time saved this week: 14-16 hours. Estimated time lost: a couple of hours and a little dignity.
Net positive. Claude, you're staying on staff.
Become an AI Advantage Member and enjoy the rest of the Newsletter
BONUS | For AI Advantage Members
Two Ways to Use Claude to Create Social Media Content (And a Prompt to Help You Know What to Post and Create the Carousel Images)...
