← All transcripts

5 Ways to Make Money From an AI the Government Fears Transcript, AI Summary & Key Points

Chris Koerner on The Koerner Office Podcast · Jun 20, 2026 · Education · 20:40 · EN

📄 Transcript

Searchable transcript of 5 Ways to Make Money From an AI the Government Fears — Chris Koerner on The Koerner Office Podcast (20:40). Search for a phrase, then click its timestamp to jump straight to that moment in the video.

Captions sourced from the original video on YouTube, published by Chris Koerner on The Koerner Office Podcast. The video, its captions and all related intellectual property remain the property of their respective owners; AINotes claims no ownership. Provided for research, accessibility and search — see the Transcript Notice and Copyright Policy.

00:00 The most powerful AI ever released to the public  was online for only 3 days. Three. And I had a chance to play around with it. But then the  United States government sent a letter and the company that built it shut it off worldwide  in a matter of hours. Not throttled down. We're talking off off. So I'm going to tell you  what it was, why the government killed it, and the stuff it did in those three days  that honestly scared the crap out of the same people that build it.

00:24 Then I'm going to  hand you five businesses that you can set up right now while this AI model is still dark. So  the day that this AI model comes back online, which it will, you are going to be primed and  ready to make some money on it. And at the end, I'm going to show you the one signal that everyone  out there is getting wrong. They're reading it backwards.

00:43 People see the word banned and they  run away. But I'm going to show you why the word banned is the strongest buy signal the government  has ever handed a regular person. Let's dive in. Okay, let's get on the same page. The company is  Anthropic. They make Claude. This is the AI that a lot of you already use, as do I. On June 9th,  they released a new model called Fable 5.

01:02 This was kind of a big deal because Fable came off of a  tier that they'd been holding back. And this tier is a class of AI models that they were calling  mythos. Think of their normal models like Opus, Sonnet, stuff that you're probably familiar  with as the strong stuff that everybody gets. But Mythos was like the locked vault in the back.  The one they said was too powerful to hand out.

01:23 So Fable 5 was anthropic finally cracking that vault  open for the public with safety guardrails bolted on. And that's important. I'm going to get back  to that. And you're probably non-technical like I am. You've probably never written a line of  code. So this matters to you because the thing that Fable can do is the exact thing that every  online business has been waiting for.

01:46 But before I get to that, I want to get to the part that  reads like a movie. Here's why they pulled Fable. So 3 days after launch on June 12th, at around  5:21 in the evening, Anthropic got a letter from the US Commerce Department. This letter ordered  them to cut off access to Fable 5 for any foreign national. But listen, that doesn't really work.

02:10 Anthropic can't just check your passport when you log in. And if they tried, that friction would  turn almost everyone away. So to follow this order from the US commerce department, they had no  choice. They had to shut down the whole thing for everybody. You, me, paying customers, enterprise,  doesn't matter. Even their own employees who weren't US citizens, they had to shut it down for.

02:31 But the craziest part about this is that this is the first time in history that the US government  used export controls on an AI model itself. Export controls, like we're talking physical goods.  This has never happened before. Here's the kind of company that this AI is now keeping. Export  controls are the rules that cover stuff like nuclear material, fighter jets, missiles, and the  guidance systems inside them, or spy satellites, night vision.

02:58 The whole list of stuff the  government decides is too dangerous to let walk out of the country. That's the club that Fable  just got drafted into. A thing that you just log into with a password is sitting on the same shelf  as weapons. But individuals have never had access to weapons or missiles or nuclear reactors until  now. It's called Fable. A couple years back, the government banned the best AI chips, the Nvidia  chips that everybody wanted.

03:17 They banned it from being sold to China. So Nvidia just turned around  and built weaker versions on purpose so they could stay legal. Of course, they don't want to stop  selling to China. So then they went after the machines that built the chips. There's this Dutch  company called ASML that makes the one machine on Earth that can actually print the most advanced  chips.

03:37 One machine costs over a 100 million bucks and it weighs like 170 tons. And the US government  said that machine does not go to China. Period. So follow the pattern. First they controlled the  chips. Then they controlled the machines that make the chips. And now for the first time ever they're  controlling the model itself, the AI. They skipped the hardware and went straight for the brain.

03:59 So  here's the part that should make like the hair on the back of your neck stand up. This is the second  time in history that the government has looked at a piece of software and called it a weapon.  The first time set off a fight that changed the internet forever. And how that fight ended is  the whole reason you should be paying attention right now, but I'm going to get to that later.

04:19 And why? Why did they do this? Well, there's the official reason and the unofficial reason.  The official reason is because of a security hole. The government said somebody found a way to  jailbreak Fable. Do you remember the early days of the iPhone? You could jailbreak it and like have  it do things that Apple didn't authorize. Someone was doing that allegedly with Fable.

04:37 And this  was like a trick to slip past its safety rules, its guardrails. And they got worried about what  a foreign actor could do with it. I mean, think about it. I I like to think about things in upside  and downside. Basically, the government was like, uh, it's kind of more downside than upside here,  at least in its current state. And that's the whole thing is what these mythos class models like  Fable are scary good at is finding weaknesses in software.

05:05 So, this is awesome if you're defending  a company, but bad if you're a bad guy and you want to break into one. Hey, please subscribe to  the channel. Whether you're on Spotify, Apple, YouTube, doesn't matter. It would mean a ton.  So, of course, Anthropic, which by the way is privately valued at about a trillion dollars.  Okay, they pushed back hard.

05:20 They said that listen, that jailbreak that you heard about was  narrow and other public AI models can already do the same thing. So yanking a model that are  used by hundreds of millions of people over one alleged jailbreak is a massive overreaction.  Now of course Anthropic is biased. Of course they're saying that. They know what this means to  their bottom line or topline both.

05:43 And the timing of this could not have been worse for Anthropic  because they're right about to go public. What just happened to SpaceX that went public? Up  40% in the first two days. We're talking a $3 trillion company. Anthropic's licking  their chops hoping for something similar. Terrible timing for them. And I don't know the  full politics behind this.

06:02 I'm sure there were some conversations that went on behind closed  doors that nobody will ever know about. But what I do know is the result. The single best  AI tool in the world went dark. It had never happened before. And the window to get ready for  it to come back swung wide open. Now we kind of know what we're getting and we can prepare for  it.

06:21 All right. Remember when I said, you know, the first time the government called a piece of  software a weapon? Here's how that ended. So, back in the '90s, there was a program called PGP,  and this program let regular people scramble their messages so nobody could read them. The government  freaked out. They classified that software as amunition, the same legal bucket as a missile, and  they opened a criminal case on the guy who built it.

06:46 He was facing literal prison time for letting  code leave the country. And how did he beat it? He simply printed the code in a book because a  physical book is protected speech. So, they were stuck. The government was stuck. Their hands were  tied. The case fell apart and they lost. And that weapon they tried so hard to lock down turned into  the little padlock that you see on every website.

07:08 It's the whole reason you're able to put your  credit card into the internet at all. So, this thing that in the '9s they called too dangerous to  export became the whole freaking foundation of the entire online economy. But let me tell you why I  care this much about this model. And again, this is not sponsored. Anthropic does not pay me, okay?  This is just Chris Kerner talking here.

07:25 Here's the one thing that makes Fable different the most  because everything else that we talk about is going to flow from that. So, every AI you've used  so far is smart for like a minute. Okay? You ask it, it answers. And if the job is big and the  context window is what it's called, gets too big, too ownorous for the model to handle, it loses  the plot.

07:47 It starts hallucinating. You're like, "Whoa, what are you talking about? Why are  you talking about this? We're over here." halfway through the chat and you're going to be  back to babysitting it. But Fable was built to be left alone, right? Like a really good employee.  You give it a real job. Let's say it has 10 steps to it and it plans it out. It does the work.

08:05 It checks its own work, fixes its own mistake, and then it hands you the finished product  with a nice little bow on top. And it can run for hours. It did. It can run for days. It can  even spin up copies of itself, split the work, and review the copies. And that's the whole game  here. The magic is not that it's just smart. Lots of AI is smart. It's all smart.

08:26 The magic is that  you can walk away and it'll just do. It'll just work while you sleep and you can trust the result  enough to put your name on it and to sell it. Okay? That's why we're here to sell it. That has  never existed before. So, if you've tried AI for real work and just watched it quit on you halfway,  you're not crazy. That wall is exactly what Fable just knocked down.

08:50 Here's some receipts. Here are  some examples of how crazy this is. So, Stripe, you know, giant payments company. They tested uh  Fable early. They had a code base with 50 million lines of code and they needed to overhaul it. They  needed to update it. And something like this would usually take a whole team of engineers 2 to 4  months. Fable did it in a freaking day.

09:05 One day. That's a real example. Another example, a physics  researcher ran it on a hard frontier problem. Now, a hard frontier problem is like a an  unsolved question in physics where our current understanding of physics just fails to get us  all the way across the finish line. The frontier problem is something that requires like a huge  leap in understanding to solve, aka a big freaking deal.

09:36 So, Fable got nearly as far in solving this  frontier problem in 36 hours as a top competing model got in 4 days. Okay, a little more gimmicky  of an example, but still interesting. Someone had it play a card game called Slay the Spire. And  when they gave it a simple memory file, so it could just take notes on its own runs of this card  game, it started winning three times more often.

10:00 So, it taught itself. It learned midtask and then  kept retraining itself from its own notes. That's freaky. That's scary. That's amazing. All of the  above. And the example that probably explains the whole reason for the ban in the first place is  this thing called a hacking skill test. Okay, this is like a pre-employment test that an  employer would give to a software developer to help them try and expose security vulnerabilities.

10:25 So the locked down twin of this model, Fable, this version did not have safety filters or  guard rails. Scored about double what the normal top model scores. Double. That reason is the main  reason why this thing has a safety version in the first place. It's it's really why the government  got nervous enough to pull it. And the most ironic thing about all this is that this mandate  locked out every foreign national.

10:49 And a country like South Korea makes a huge percentage of the  memory chips that power these American AI models in the first place. So we cut off the exact people  building the parts, which I just find hilarious. So, the part that you, you know, you probably  came here for is the money. I want you to do something right now. Open your bank statement and  add up every single software subscription that you're paying for.

11:14 Your CRM, your email marketing  tool, text message platform, calendar booking app, funnel builder, review management, even a separate  form builder or automation tool. I've talked to business owners paying $500 to $3,000 a month  across eight or 10 different tools, and half of them don't even talk to each other. So, then  they go pay for Zapier on top of that just to make things connect.

11:34 High Level replaces all of it.  CRM, email, SMS, funnels, websites, calendars, automations, reputation management, invoicing,  one login, one dashboard, one bill starting at 97 bucks a month. A guy in my community was paying  3.47 just for ClickFunnels, 99 for Mailchimp, 25 for Calendarly, 79 for a review tool, and 49  bucks for Zapier just to glue it all together.

11:57 That's 600 a month if you're not counting.  He moved everything to high level and cut his software cost by over 80%. But the best part  isn't even the savings. It's that everything just works together natively. A lead fills out a form  on your funnel that creates a contact in your CRM, which triggers a text and email sequence, books  an appointment on your calendar, and then ask for a review when the job's done.

12:15 One workflow, zero  duct tape, and then you can white label the whole platform to resell it to other businesses as your  own software. Charge them 2 to 500 bucks a month. Check out gohighlevel.com/tkopod. They'll give  you a 30-day free trial to see what I mean. Here's five ideas of businesses that can be built on this  one superpower. And for each example, I'm going to show you a real business that's already making  money the hard way with employees.

12:36 So, like these are not like wacky off-the-wall ideas. These are  already proven to be working with humans the hard way, the old school way, right? The 2025 way.  Okay. Idea number one, legacy code rescue. So, there are agencies right now that charge 30 to 70  grand to move one old clunky software system over to something modern. Okay, this is a big problem  that companies need solved.

13:00 And one job could run into the hundreds of thousands easily cuz a  company might have hundreds or thousands of employees and it's just very tedious. And this one  job can take a team of engineers like 6 months to complete. So, they're justified in charging  hundreds of thousands of dollars. But Fable could do this entire migration on its own. So to  find customers, instead of cold calling owners, use tools like BuiltWith or Outscraper or a  freaking Chamber of Commerce website to find businesses that are still running

13:32 on ancient  systems. Then you want to talk to the office manager because that's the person who probably  hates that old ancient system the most and just say, "Listen, I can get you off your old system in  a weekend with zero downtime." The companies that do this today are called modernization firms.  By the way, there's a pretty good chance that you watching that you, the person consuming this  episode right now, work for a company that uses an ancient system like this.

13:57 Who's saying that  that company, the one you work for today, can't be your first customer? Did you know that 95% of  ATM transactions are run on an outdated software system called Cobalt? Ancient. Tons of million,  hundred million, billion dollar companies are running on 20-year-old software. You don't have  to start with a big Fortune 500 company. In fact, I'd advise you to not start with a local company  that has 13 employees and needs to migrate from QuickBooks Desktop to QuickBooks Online.

14:26 Something  simple. You won't be able to charge nearly as much, but you can at least get your feet wet. Idea  number two, the overnight turnaround shop. Okay, there's a ton of small to medium-sized  businesses that are way behind on their books, their financials, and they end up having to pay  between 3500 to 10 grand just to get their books cleaned up.

14:45 And this is a miserable, tedious  job, which is why few people want to do it. Now, using today's AI models, before Fable, you'd  still have to babysit these transactions and keep training and retraining them and worry that  your context window runs out. But with Fable, you could just drop the mess in everything at 5:00  p.m. all the account statements, bank statements, spreadsheets, everything, and it will reconcile  everything overnight by itself.

15:07 and you can review and deliver it to your customer by 9:00 a.m. Go  to bed, wake up, it's done. You know how many Facebook groups there are for bookkeepers that  are just turning down what they call cleanup work because they hate it and they have more important  things to do. You don't need to be a bookkeeper. You don't need to have a finance background.

15:26 Post  in these groups that you take overflow cleanup and you have 48 hour turnaround, which is unheard of.  You do one of these a week, three grand a piece and six figures a year. Your costs are almost  nothing. All right, number three. You remember my episode with Brett Williams, the design joy guy,  does like 60 to 80k a month with a design studio, one person design studio.

15:46 He's charging five grand  a month per client and he's capped out cuz he's the one doing all the designing. But Fable could  be the team that he never had to hire. It could be your team, your agency team. So find a service. Go  back and listen to that episode. We'll link to it below. Productize it. People can have one request  in at a time. Make Fable do the work.

16:01 Number four, how about a custom product for every single  customer? So, we all know that a generic sales email is not nearly as good as a personalized  one. Your reply rate will jump 2 to 3x, but companies aren't really doing this for sales  emails cuz with a human, it takes too long. And there are AI tools that will kind of do this  for you, but they're not super accurate.

16:23 I get these emails every day. But with every piece of  personalization in your email, your reply rate will go up by a few percent, right? First name.  Great. First name plus business that they own. Even better. First name plus business that they  own plus city they live in. Awesome. Why not 8 to 12 pieces of personalized information? Fable can  do that.

16:42 And if you don't want to send these cold emails to acquire customers, then you just find  a customer, a business customer that has an email list that needs to get more out of it. And almost  every business is not fully utilizing their email list or even close to it. So you take their list,  their strategy, their process, attach Fable to it, and you can make amazing things happen with their  email list.

17:04 Doesn't need to be a big company. It could be a gym, a med spa, anything. There's no  business too small for this as long as they have emails. Number five, how about digital assets that  run themselves? So, there are people out there, I've interviewed a lot of them, that have little  portfolios of small websites that earn, you know, 300 bucks to 3,000 bucks a month each, but they  have multiple or dozens of these little digital assets, and it really adds up over time, like  directory sites, right?

17:31 Being a website landlord. Now, these portfolios can be a grind to build  and to feed, but Fable can build them and keep improving the winners on its own for days. It can  log in to Google Analytics, Google Tag Manager, Google Search Console. It could see where the  holes are and automatically fix them. It's going to make a lot of mistakes for sure. It's going  to miss a lot of things for sure.

17:51 You're going to have to train it. But the training process is  so much shorter now. And the work ethic from this model is unlike anything we've ever seen. So, what  can we do now while Fable 5 is still dark? Don't just sit around and wait. Pick one of these five  ideas. Or better yet, pick one of your own cuz you're probably going to love that one even more.

18:11 Pick one niche, pick one offer, one landing page, whatever, and start using the inferior models,  which are still incredible. I like Opus 4.8 to start hacking together some of these things  and just know that when Fable 5 does come back, it's just going to be easier. But here's the  truth of it. If this model got switched off once, it can get switched off again.

18:31 It can get nerfed.  So don't bet your entire life on one AI tool. You always want to test a bunch of different tools.  It's going to give you more insight into each one of them individually because you'll have something  to compare them against. The AI should be the engine or the gasoline. It's not the car. It's not  the company. So, the signal that I referred to at the beginning that everyone's reading backwards is  when the government wants to slow something down, they regulate it.

18:56 We've seen it over and  over again. They do it with military drones, with nuclear materials, fighter jets, missiles,  supercomputers. When they think something can shift the balance of power in the world, they  export control it. Same as a weapon. They just did it to a piece of software that you can rent  for 10 bucks. The most powerful institution on the planet looked at this tool and said,  "Uh, this is as dangerous as a missile."

19:21 But individuals have never had access to things as  powerful or dangerous as missiles until now. So, don't hear the word banned and think, "Oh,  stay away." The truth is the opposite. The US government just did all of your market research  for you. They told you in writing that this is the most powerful software ever built. So the only  question left is whether you spend this blackout getting ready or whether you find out like 2  years later like everyone else will.

19:42 By the way, I have a new private YouTube community where I  post exclusive interviews and videos that no one else in the world will see. All you got to do  is pay five bucks a month. And you know what? YouTube takes 30% of that. So I'm hardly getting  anything from that. In addition, you're going to get every single 25 plus page business plan that  we make from every episode we publish.

20:00 And we sell these for 19 bucks and we do 15 episodes a month.  So, this is about $300 a month worth of value, plus extra interviews and videos that you'll  never see anywhere else for five bucks a month, and you can cancel anytime. You'll also get my  previous backlog of every business plan I've ever published. So, you can get the link right  down below.

20:22 It's all done right here on YouTube. And you'll even get little custom emojis or stars  next to your name when you comment on videos so you can get priority replies from yours truly.  Thanks a lot. If you like this content, please share with a friend, hit a like or a comment  below, and we'll see you next time on the Kerner

💡 Answer

Prepare businesses that use AI for legacy-code migration, overnight bookkeeping cleanup, productized services, personalized customer outreach, or self-running digital assets before Fable 5 returns.

🧠 AI Summary

Fable 5, an AI model released by Anthropic, was shut down worldwide after the US Commerce Department ordered Anthropic to restrict access for foreign nationals, reportedly because of a jailbreak and the model's exceptional ability to find software vulnerabilities. Fable 5 could perform long, multi-step tasks autonomously, check and correct its work, run for days, and divide work among copies of itself. The blackout creates preparation opportunities in legacy-code migration, bookkeeping cleanup, productized services, personalized sales, and self-running digital assets. The central recommendation is to build a business around a customer problem rather than depend on one AI model, while testing multiple tools.

🔑 Key Points

  • Anthropic released Fable 5 on June 9 and shut it down worldwide three days later after receiving a June 12 letter from the US Commerce Department.
  • The order targeted access by foreign nationals, but Anthropic shut down Fable 5 globally because it could not practically verify users' nationalities without severe login friction.
  • Fable 5 was designed to plan and execute long tasks, verify its work, fix errors, run for hours or days, and create copies of itself to divide and review work.
  • Stripe reportedly used Fable to overhaul a 50-million-line codebase in one day, compared with a typical 2 to 4 months for a team of engineers.
  • The five proposed businesses are legacy-code rescue, overnight bookkeeping cleanup, productized services, personalized customer outreach, and self-running digital assets.
  • Businesses should begin with current AI tools, choose one niche and offer, and avoid depending entirely on Fable 5 because it could be shut down or weakened again.
  • Government export controls are presented as a signal that a technology is considered strategically powerful, but the transcript does not establish that Fable 5 will return.

✅ Actionable items

  • Add up all software subscriptions and identify businesses paying for disconnected CRM, email, SMS, funnel, calendar, review, form, and automation tools.
  • Choose one niche, one offer, and one landing page, then prototype the business with currently available AI models.
  • Find companies using legacy systems through BuiltWith, Outscraper, or a Chamber of Commerce website, and approach the office manager about migration work.
  • Start with a smaller migration, such as moving a 13-person local company from QuickBooks Desktop to QuickBooks Online, before pursuing larger projects.
  • Advertise 48-hour overflow bookkeeping cleanup in bookkeeping Facebook groups.
  • Productize a service with one customer request active at a time and use AI to perform the work.
  • Use a customer's email list, strategy, and process with AI to create more personalized campaigns for businesses such as gyms or medical spas.
  • Build and maintain portfolios of small websites, using Google Analytics, Google Tag Manager, and Google Search Console to identify and fix weaknesses.
  • Test multiple AI tools rather than basing an entire business on one model.

💡 Business ideas

Legacy code rescue12:52

Migrate outdated company software systems to modern systems using Fable 5.

For
Businesses running ancient software systems, beginning with smaller local companies.
Solves
Expensive, tedious, and lengthy legacy-system migration.
Validate by
Find companies using old systems through BuiltWith, Outscraper, or a Chamber of Commerce website and approach the office manager.
  • Moving a local company from QuickBooks Desktop to QuickBooks Online.
  • Modernization firms charging 30 to 70 grand for one migration.
Overnight bookkeeping cleanup14:34

Reconcile overdue books and financial records overnight with AI.

For
Small to medium-sized businesses and bookkeepers with overflow cleanup work.
Solves
Tedious, backlogged bookkeeping cleanup.
Validate by
Post in bookkeeping Facebook groups offering overflow cleanup with a 48-hour turnaround.
  • Drop account statements, bank statements, and spreadsheets into the system at 5:00 p.m. and review the output by 9:00 a.m.
Productized AI service agency15:40

Use Fable 5 to deliver a repeatable service that previously required a larger team.

For
Businesses that need the selected service.
Solves
A solo operator's capacity is limited by the amount of work one person can perform.
Validate by
Choose and productize one service, then limit customers to one active request at a time.
  • Design services modeled on Brett Williams's one-person design studio.
Personalized customer outreach16:06

Create highly personalized sales emails or campaigns for businesses using their existing email lists.

For
Businesses with email lists, including gyms and medical spas.
Solves
Human-created personalization is slow, and generic outreach produces weaker replies.
Validate by
Work with a business that has an underused email list and apply AI to its existing strategy and process.
  • Personalizing messages with first name, business, city, and 8 to 12 pieces of information.
Self-running digital assets17:14

Build and improve portfolios of small revenue-generating websites with autonomous AI work.

For
The operator of a portfolio of small websites.
Solves
Building and continually feeding and improving many websites is labor-intensive.
Validate by
Start a small portfolio and use analytics and search tools to identify performance gaps.
  • Directory sites earning 300 bucks to 3,000 bucks a month each.

🏗️ Business models

White-label software resale12:14

Use High Level as an integrated business-management platform and resell a branded version to other businesses.

  1. Use the platform's CRM, email, SMS, funnels, websites, calendars, automations, reputation management, and invoicing.
  2. White-label the platform.
  3. Resell it to other businesses under the reseller's own brand.
  • A form submission creates a CRM contact, triggers text and email sequences, books a calendar appointment, and requests a review after the job.
AI-enabled service agency16:00

Productize a service and use Fable 5 as an autonomous production team instead of hiring a large employee team.

  1. Choose a service.
  2. Productize the service.
  3. Allow one customer request at a time.
  4. Use Fable to perform the work.
  • A one-person design studio could use Fable as the team it never hired.
Digital-asset portfolio17:14

Own multiple small websites that earn recurring revenue and use AI to build, maintain, and improve them.

  1. Build a portfolio of small websites such as directory sites.
  2. Monitor performance using Google Analytics, Google Tag Manager, and Google Search Console.
  3. Fix weaknesses and improve winning sites.
  • Being a website landlord.

💰 Monetization

High Level platform subscription Starting at 97 bucks a month. 11:38

High Level offers a combined software dashboard for businesses.

  • CRM, email, SMS, funnels, websites, calendars, automations, reputation management, and invoicing.
White-label software resale Charge them 2 to 500 bucks a month. 12:20

Resell a white-labeled version of High Level to other businesses.

    Legacy-system migration fees Agencies charge 30 to 70 grand for one old-system migration; larger jobs can run into the hundreds of thousands. 12:52

    Charge companies for moving old software systems to modern platforms.

      Bookkeeping cleanup fees Between 3500 to 10 grand per cleanup; the example proposes three grand per job. 14:38

      Charge businesses for cleaning up overdue books and financial records.

        Productized service retainers Brett Williams is described as charging five grand a month per client. 15:45

        Sell a repeatable service package to clients while using AI to perform the work.

          Digital-asset revenue 300 bucks to 3,000 bucks a month each. 17:14

          Earn recurring income from portfolios of small websites.

          • Directory sites.

          📣 Marketing

          Sales

          • Offer legacy-system migration with a weekend turnaround and zero downtime.
          • Offer bookkeeping cleanup with a 48-hour turnaround.
          • Start with a smaller local customer to gain experience before pursuing large migrations.

          Branding

          • White-label High Level as the reseller's own software.

          Distribution

          • Use personalized cold emails with 8 to 12 pieces of information.
          • Apply AI to a customer's existing email list, strategy, and process.
          • Use a landing page for a focused niche and offer.

          Customer acquisition

          • Use BuiltWith, Outscraper, or a Chamber of Commerce website to identify businesses using outdated systems.
          • Approach the office manager at a company with an old system.
          • Post an overflow-cleanup offer in bookkeeping Facebook groups.
          • Find businesses with underused email lists, including gyms and medical spas.

          🔍 SEO & discoverability

          Mistakes

          • Relying on one AI tool as the entire business.
          • Using generic sales emails instead of personalized outreach.
          • Starting with an overly large or complex migration instead of a smaller local project.

          Strategies

          • Use Google Analytics, Google Tag Manager, and Google Search Console to find weaknesses in digital assets.

          Other channels

          • BuiltWith, Outscraper, Chamber of Commerce websites, bookkeeping Facebook groups, and email lists.

          🧭 Frameworks

          Five AI-enabled business opportunities12:52
          1. Legacy code rescue
          2. Overnight bookkeeping cleanup
          3. Productized service agency
          4. Personalized customer outreach
          5. Self-running digital assets

          🧰 Tools & AI usage

          • Fable 5 — Autonomous execution of complex, long-running tasks.00:56
          • Opus 4.8 — Prototype business ideas while Fable 5 is unavailable.18:17
          • High Level — Combine business software functions in one dashboard and enable white-label resale.11:38
          • BuiltWith — Find businesses using outdated systems.13:26
          • Outscraper — Find businesses using outdated systems.13:26
          • Google Analytics — Monitor digital-asset performance and identify weaknesses.17:47
          • Google Tag Manager — Inspect and manage digital-asset implementation.17:47
          • Google Search Console — Identify search-related weaknesses in digital assets.17:47

          AI is used for

          • Long multi-step work — Plan, execute, check, and correct complex jobs autonomously.07:59
          • Software vulnerability discovery — Find weaknesses for defensive security work or potentially enable attacks.04:53
          • Software migration — Overhaul and modernize large legacy codebases.08:55
          • Bookkeeping reconciliation — Clean up account statements, bank statements, and spreadsheets overnight.15:00
          • Personalized outreach — Generate highly personalized sales emails and campaigns.16:13
          • Website portfolio management — Build, monitor, and improve digital assets using analytics and search data.17:14

          📊 Numbers mentioned

          Costs

          • A cited software stack costs 3.47 for ClickFunnels, 99 for Mailchimp, 25 for Calendarly, 79 for a review tool, and 49 bucks for Zapier.
          • Moving that stack to High Level reportedly reduced software costs by over 80%.

          Growth

          • SpaceX is described as rising 40% in its first two days after going public.
          • Fable solved nearly as much of a frontier physics problem in 36 hours as a competing model did in 4 days.
          • Fable's version without safety filters scored about double the normal top model on a hacking skill test.
          • A card-game agent won three times more often after using a simple memory file.

          Pricing

          • High Level starts at 97 bucks a month.
          • White-labeled High Level software can be charged at 2 to 500 bucks a month.
          • Legacy-code agencies charge 30 to 70 grand per migration.
          • Bookkeeping cleanup costs between 3500 to 10 grand.
          • Personalized sales emails are described as producing a 2 to 3x reply-rate increase.
          • The private YouTube community costs five bucks a month.

          Revenue

          • Businesses are described as paying $500 to $3,000 a month across eight or 10 software tools.
          • One bookkeeping cleanup per week at three grand per job is described as producing six figures a year.
          • Brett Williams's design studio is described as earning 60 to 80k a month.

          ⚖️ Advantages, risks & lessons

          Advantages

          • Fable 5 can operate for hours or days without continuous supervision.
          • It can plan tasks, check its own work, correct mistakes, and divide work among copies of itself.
          • The model could reduce the labor required for software migration, bookkeeping, design, outreach, and website management.
          • High Level integrates multiple business functions natively in one dashboard.

          Risks

          • Fable 5 was shut down worldwide after a US Commerce Department order.
          • The model could be shut down or weakened again.
          • Fable 5 can make mistakes and miss important details.
          • Its capability to find software weaknesses could be dangerous if misused.
          • Dependence on one AI tool creates business continuity risk.

          Lessons

          • Build the company around the customer problem and offer, not around one AI model.
          • Use the blackout period to prepare and prototype.
          • Start with a manageable niche and offer.
          • Compare multiple AI tools instead of relying on one.
          • Government regulation can signal that a technology has significant strategic power.

          💬 Quotes

          The AI should be the engine or the gasoline. It's not the car. It's not the company.

          Concise principle for building an AI-enabled business without depending on one model.18:46

          The US government just did all of your market research for you.

          Summarizes the video's interpretation of export controls as a signal of Fable 5's importance.19:31

          📈 Investment analysis

          bullish

          Price references

          AssetPriceTypeTimeframeContext
          Anthropic about a trillion dollars private valuation Anthropic is described as privately valued at about a trillion dollars and preparing to go public. 05:20
          SpaceX $3 trillion company valuation SpaceX is described as a $3 trillion company that went public and rose 40% in its first two days. 05:48

          Predictions

          • return Fable 5 — Fable 5 will come back online.
            The speaker states that the model will return after the blackout.
            00:29

          Actions noted

          • Prepare businesses during the blackout rather than treating the ban as a reason to avoid the technology. Fable 5
            Use multiple AI tools and do not depend entirely on Fable 5.
            18:21

          Market factors

          • US government export controls — They can restrict access to strategically powerful technologies and may create a preparation window for businesses. 18:44
          • AI model shutdown or weakening — A business dependent on one AI tool could be disrupted. 18:31

          👤 People & companies

          Chris Kerner

          The speaker identifies himself by this name.

          07:25
          Brett Williams

          Guest associated with Design Joy, described as running a one-person design studio earning 60 to 80k a month.

          15:40
          Anthropic

          Company that makes Claude and released Fable 5.

          00:56
          Claude

          AI product made by Anthropic.

          00:56
          Nvidia

          Chip company whose AI chips were reportedly restricted from being sold to China.

          03:17
          ASML

          Dutch company described as making the machine used to print the most advanced chips.

          03:32
          Stripe

          Payments company that reportedly tested Fable on a 50-million-line codebase.

          08:55
          SpaceX

          Company described as having gone public and rising 40% in its first two days.

          05:48
          High Level

          Software platform combining CRM, email, SMS, funnels, websites, calendars, automations, reputation management, and invoicing.

          11:20
          Zapier

          Automation tool mentioned as a way businesses connect separate software tools.

          11:22
          ClickFunnels

          Funnel software referenced in a software-cost example.

          11:50
          Mailchimp

          Email marketing software referenced in a software-cost example.

          11:50
          Calendarly

          Calendar tool referenced in a software-cost example.

          11:50
          BuiltWith

          Tool recommended for finding businesses using old systems.

          13:26
          Outscraper

          Tool recommended for finding businesses using old systems.

          13:26
          Google

          Company whose Analytics, Tag Manager, and Search Console products are mentioned for managing digital assets.

          17:47

          🔗 Links mentioned