The Startup Ideas Podcast - You are using Claude Fable 5 wrong

Episode Date: June 11, 2026

Get my Fable 5 prompt pack: https://startup-ideas-pod.link/fable5-prompt-pack In this episode I break down how to get the most out of Fable 5, the most powerful model I've ever used. I move past the benchmarks and go straight into tactical use cases, copy-and-paste prompts, and startup ideas you can build today. I walk through tournaments for copy and landing pages, an interview-before-build workflow that hunts for product-market fit, and ways to point Fable at contracts, churn data, and years of your own notes. I close with three of my favorite startup ideas — a synthetic focus group firm, 48-hour custom software, and a contract refund firm — plus the exact prompts behind each. My goal here stays simple: leave you ready to build and earn with Fable 5 while it remains included in your plan. Timestamps 00:00 – Intro 02:22 – Anthropic Employee Edits a Launch Video With Fable 05:50 – Building an AI Content Engine 07:30 – Best way to configure Fable 5 08:42 – Prompt 1: Copywriting Tournament for Landing Pages 13:18 – Prompt 2: The Interview-Before-Build Prompt 18:34 – Prompt 3: Hire Fable to Kill Your Company 20:18 – Prompt 4: Your One-Page Operating Manual 21:20 – Prompt 5: Find the Gaps Worth Filling 22:06 – Prompt 6: Negotiation Simulator 23:11 – Prompt 7: The 80-Page Second Opinion on Contracts 24:56 – Prompt 8: Make Fable Build Its Own Tools 25:47 – Startup Ideas 31:23 – Closing Thoughts Key Points I show why low effort is the alpha, since Fable Low beats Opus High on routine work. I run tournaments — landing pages and ad copy scored by AI judge panels — to ship far stronger output. I use an interview-before-build prompt so Fable pushes back and writes specs with real product-market-fit odds. I point Fable at big datasets — contracts, churn data, support tickets, years of notes — to surface money and patterns. I share startup ideas Fable 5 makes viable today, including a synthetic focus group firm and a contract refund firm. The #1 tool to find startup ideas/trends - https://www.ideabrowser.com LCA helps Fortune 500s and fast-growing startups build their future - from Warner Music to Fortnite to Dropbox. We turn 'what if' into reality with AI, apps, and next-gen products https://latecheckout.agency/ The Vibe Marketer - Resources for people into vibe marketing/marketing with AI: https://www.thevibemarketer.com/ FIND ME ON SOCIAL X/Twitter: https://twitter.com/gregisenberg Instagram: https://instagram.com/gregisenberg/ LinkedIn: https://www.linkedin.com/in/gisenberg/

I've got a dumb question. If the world's most powerful AI model just came out, Fabel 5, how are we supposed to use it to make money and be more productive? I mean, I've seen some crazy examples on X of people one-chotting like a monopoly game, and that's really cool. But how could I use this super intelligence and just point it to different problems to make me money and be more productive? That's what this episode is about.

By the end of this episode, you're going to understand why Fabel 5 really matters. But more importantly than that, you're going to understand practically how you can use Fable 5 to make money and be more productive. Now is the time that you have this unfair advantage for actually pointing this thing, this insanely powerful model at problems that could build businesses that could make you more productive. And to be honest, it's a lot of fun. So enjoy the episode and I'll see you in there. Oh, I needed to make this video. I needed to make this video because 99% of people are using Fable 5 wrong. I mean, this is the most powerful model that we've ever seen.

And people are just, they're just using it the wrong way. They're not seeing all the opportunities, all the ways to make money, be more productive with it. So I had to make this episode where I break down how you're using Fable 5 wrong. What are a bunch of use cases and prompts that you can use today that's going to help you? and a bunch of startup ideas that now, with the most powerful coding agent ever available to mankind, what has that unlocked? That's what this episode is all about.

This is a huge unlock, yet all the episodes that I've seen and tutorials of videos that I've seen on the internet have been about benchmarks and things like that. But I wanted to get to the tactical use cases for Fable 5 so that you, after this video, if you stick around to the end, you understand how to use Fable 5 in terms of use cases and in terms of startup ideas so that you can actually build a business and start making money with Fable 5.

Let's get right into it. The first thing is this guy who works at Anthropic used Fable to edit and launch his own launch video. And it looks spectacular. I can't believe how good it is. And I will say he definitely had a good camera. You can see here. He, you know, he's got a good setup. He's got this cool bookshelf. But he did 17 takes and four scenes. And it ended up with this beautiful visual, right? And you can see the text looks amazing. This is something that would have cost hundreds of thousands of dollars to make. make and using Fable with two video edit is totally possible. The prompt is here and I'm going to read it to you. I'm processing the recording of a bunch of videos in Fable Full recording. The script for them is in Fable Full Recording, Fable 5 script. I want you to run the 11 Labs transcription service on them and stitch together the best shots into one final clip.

A few notes, there are multiple takes per scene. The best takes are usually the one at the end with the fewest ums, but that can change. I also reshot the first scene at the end. They should all be organized by timeline. For a few of them, I start the video by saying, hey, name, as a way of starting the sentence, warm. We should cut that out. Create a JSON file.

Use FMMPEG. Orchestrate this using workflows slash goal. Don't stop until you have the final video. So what do we learn from here? We know that he used 11 labs. He stitched together the best shots. They created a JSON file. He used FMPEG.

And he orchestrated all of this using workflows using slash goal and said, don't stop until you have a final video. And it worked. You know, I can go through this deck. But you can see here, he uses whisper, whisper flow. It removed all the ums. The subagents pick the best takes. The edit is in this JSON file.

And then FFMPEG executes the JSON. It stitches together the first cut. from the prompt, create a final scene using FFMPEG. It actually color grades with this prompt. The color grading feels a bit muted. Can you fix that? It's able to color grade. And it looks amazing.

Look at that. Just absolutely perfection. That's beautiful. And anyone who's ever done any color grading before knows how hard this is. The static design frames. are all there. He uses re-motion to stitch it altogether. Each PNG is rebuilt as code. And then there's one file that sets the feel. So you've got this, you know, beautiful, beautiful one file here. And the point here is you can create a beautiful, a darn beautiful video.

You can export it to Figma. You can use the Figma MCP. and you can just use Fable to build something beautiful. So many people, so many of us are vibe coding apps, can't get people to those apps. But when you have something like Fable now, you can use it to actually grow your whatever it is you're working on. And this is a great example of that. That's one example.

The two is, and this is, I saw a WIS of Ecom tweeted this, you can create an AI content engine. I saw other people basically asking Fable to create tweets, and they were saying, it's not actually not that good at writing. It's not really good at writing. It's not good at writing because you haven't put the right inputs,

the right strategy, the right weekly engine, and then connect it to Fable. So if you want to use Fable as a content engine, here's how you can do it. You're going to want to put your inputs, so you're going to want your origin story, you're known for, your offer, your ICP, your frameworks,

and you can ask Fable to actually help you with that. You can put that in your content brain. You can use something like a obsidian here, your pillars, your funnel structures, your tone. Then every week you can do this research, scanning niches, finding topics, performance, three hypothesis, test weekly, and that creates the assets.

This is this weekly loop, and that uses Fable 5. So Fable 5 runs every system. reads your charts. One of the great things about Fable 5 is it's able to visually check out using co-work. You know, browse are using co-work. It's able to go and see things and really, really understand it, and it works for hours on its own. And the third piece of thing that I think people are getting wrong with Fable. And then I want to get into actually going into Claude, showing some use cases, and giving startup.

ideas is my friend Morgan Linton said this. He says, my biggest takeaway from my first day with Fable, low effort is the alpha. So the mistake a lot of people are making is, you know, the defaults on high. But I think that as, you know, I think it's June 22nd, it's going to be API pricing. It's not going to be included in your Macs account. You're not, you know, you're going to want to obviously not spend a ton of token. So how do you do that? You're going to want to use something like factory.AI. They have a product called Droid, and Droid is going to basically orchestrate when you're using Fable, what effort you're using at, and when are you using something like Opus or

other models that are more routine tasks. So I think you're going to start to see just more people using things like Droid, more people actually doing lower effort for more routine tasks, and just so you know, someone tweeted this here. In fact, Fable low is better than Opus High, so it's pretty. impressive. So if people want me to do a full breakdown on how to use droid and factory AI to set up this orchestration, just comment, I can do it, and I'll show you how you can set that up. All right, let's talk about use cases.

Let's start with this one, actually. If you're anything like me, writing copy is really difficult. And one thing I'm learning is that, you know, copy, uh, One thing I'm actually relearning is that if you have good copy, you're going to be able to, your conversion rate's going to be higher. So what's really cool is you can use, you can actually set up a copywriting tournament so that you end up creating absolutely stunning websites, building websites, using Fable with incredible copy. So look at this one prompt that I got for this SaaS idea. I took the SaaS idea from ideabrowser.com. I screenshot I put it in and I was able to get this incredible idea around backyard homes. So you put in your address.

It checks your lot and says, hey, can you build these ADUs, which are basically mini houses in your backyard? And when you look at this copy, can you build a backyard home on your lot? enter your address. In minutes you'll have a written verdict. This is a real report. Yours looks like this and it actually creates it. Look how beautiful this is.

The little animations. The copy is incredible. I'm going to show you how to do this. And the way I do it is with something called a landing page tournament. So let's go into the prompt. By the way, in this episode, I'll include all prompts in the description. And so you can go and grab them. So I say, write my landing page eight different ways, different hook, different structure each time,

then create five judges, a skeptical CFO, a distracted founder scrolling at midnight, a competitor, my ideal customer, and a conversion copywriter. Have every judge score every version and explain the scores. Kill the losers, merge what worked into a final version, and show me the scoreboard so I can see why it won. And then I just basically attached the idea of the day, can I build a backyard home tool to find in minutes from ideabrowser.com? I put that in there just to see if it would work. And let's actually look at what happened. So what's great about doing these tournaments, so I use Fable 5 high over here, is it basically, instead of getting one option that maybe is okay, you're basically getting it, you're pitting it against each other. you're getting just a lot better, a lot better output. So when we go to the landing page

tournament MD, we basically got eight versions of copy. We had five judges, 40 scores, and one winner. And it just does such a good job. I'll go through one of these and then I'll go to the end just so you can get a sense. So actually, I'm just going to skip to the end here, just so you can see what happened. So it created a scoreboard final version. So the CFO scored at this, the founder scored at this, the competitor scored at that. And then based on that, it said this is just a lot better. So I have been writing copy for a lot of years on websites and apps. I get millions, I have millions of users.

And when I saw the backyard copy, like the one that won, I was like, this is incredible. And the secret sauce is that I use this framework. You can see here, like the CFO says, yeah, I'm going to score this a six because 2000 a month is a claim the page must immediately defend or it reads like a Zillow estimate. This is what's so crazy about Fable 5 is it's scary, it's scary good. Like, yeah, a CFO might say that. the midnight founder who's scrolling in the middle of night reads $100,000 on a guess is a gut punch and it stops the scroll.

It's just, it's so good. And this is just a really, it's something that I'm seeing no one, no one use that I think you should be using as well. Another thing that Fable 5 is just incredible at is plan. This is something that if you're building a startup, it's worth considering. So the interview before the build. So the prompt goes, I want you to build this rough idea. Don't build anything yet.

Interview me like Mark Zuckerberg or Sam Altman or Brian Chesky, experts at building companies that get PMF. One question at time, 15 questions max, hunting for the thing I haven't thought about. Push back when my answers are vague. after the interview, write the full spec, and list three ways this fails, then build the V1. So what I'm seeing on X and other places is people are just one prompting these startup ideas. And they're not getting really what they want. It might look visually great, but from my perspective, I don't think it's something that is a high likelihood of PMF, product market fit.

So we're actually going to use this prompt to, I'm going to go scroll up. We're going to use this prompt so that we can have a spec, a full spec, that has the highest likelihood of success. So I grabbed the idea of the day from ideabrowser.com. It's basically this habit widget that pings you at the right moment. Imagine a habit at that basically gives you a push notification and says, hey, you need to work out or you need to drink, water or something like that. And what's so crazy about Fable 5 high in this case is it asked me questions. We're going to go through some of the questions. It says, what habit have you personally failed to keep in the last 12 months? And I said, failed workout treats. And then I said,

and I'm just a person who wants to unlock my potential and can always do better. And then it goes, that's half an answer. Like, I'm getting pushback. from Fable. Fable feels like a real person. If that was a real person, if that was a Brian Chesky, if that was Sam Altman,

and they were actually trying to help me build an app that gets to PMF, they would say something like that. And to this date, I have never had an LLM speak to me like that. And I absolutely loved it. Because that's what I need

in order to create an app that has a high likelihood. high likelihood of success. So it says, walk me through the actual death of your last workout streak. What were you supposed to work out? What time of day?

What happened on the day you missed? Did you miss because you forgot or because you remembered and chose not to? This matters more than anything in that screenshot. The entire thesis is timing is the problem. A smarter ping fixes it. And then I said, I let life get in the way. Then it goes,

that's a horoscope. answer Greg. Life got in the way is what everyone says when they haven't watched the tape. You never let a podcast guest get away with that and you'd say, give me the scene. You love to see it. It pushes back and it helps me, you know, I'm not going to go through all of this because that's not the point. The point is it pushes back on you. And by the end of this, by the end of this, we got an incredible, incredible outcome. And I want to, I want to, I want to talk about it. It says the interview changed the product more than you might realize.

What we're building shares a name with the idea browser write-up and almost nothing else. No widget. Widgets can't ping you. No week of manual check-ins. You banned homework. No habit platform workouts only. It's a phone number that can see your calendar and text you with receipts and has a consequence you signed up for. So basically what it did is it took this idea of like you're going to get a text message.

Well, you're going to connect it to your calendar. and it's going to basically say, you should work out now. And if you don't, you're going to get a consequence. It built the app. Completely works. It built the spec document.

And it's just something that has a high likelihood of success. So I don't see anyone doing this. Using Fable 5, helping you interview, you know, the key is say it's a Mark Zuckerberg. say it's a Belz, a Brian Chesky. And then you get this spec document that really just figures out who you are, what the onboarding should be, what the loop should be, and just gives you the highest likelihood of success,

and gives you the architecture. You can see here it says build with Python, fast API, AP scheduler, what the pricing should be, and what should be out of scope for V1? So just an incredible output, and I've never seen an LM do something so high quality. We're going to go through a few more use cases. I'm just going to just lightning around it. I'm not going to show it in Clod.

But it's just to help you get your creative juices flowing for, okay, how else can I use this thing? Well, you can hire it to kill your company. So everyone asks for AI, everyone asks AI for growth ideas, but you can flip it. Give it everything and a day to destroy you. The magic is, it ranks threats by which one could execute itself, which is the only threat ranking that matters now. So here, check this out. You say, here's my P&L. Here's my pricing page.

My turn data. My last 50 support tickets. You are a well-funded founder who hates me. Spend the day building the company that kills mine. Real plan. Positioning, pricing, first 10 customers you'd steal and the exact email you'd send each one. Then rank every attack by how much of it you can do yourself with no employees. Don't be polite.

I'd rather hear it from you than read it in my mind in the churn dashboard. The output of this is incredible. So the insight here is you want to mine your support tickets. Fable is incredible with big data sets. it can look at your churn data and it's going to help you make more money from a SaaS. Now you can also go and buy a SaaS now. You can actually go and refactor that code using Fable.

You can go and look at those support tickets. I've talked about this on the channel. There's an opportunity, especially now with Fable 5, to actually go buy things for not a lot of money, kind of flip it, you know, or renovate it, maybe flip it or just keep it for profit. and cash flow, but the idea is basically now you have something that is just so high quality. We talked about the copy tournament. This is a really interesting one. So you give it two years

of your notes, your decision docs, your post-mortems, you ask it to map every meaningful decision, what I chose, what I believed at the time, what actually happened, find my patterns, where am I systematically too early, too late, too optimistic, What do I always say right before a bad call? Write me the one page. Operating manual for working Greg that a CEO would secretly keep. So what's really cool about Fable 5 and where we are today, 2026, is you've got a $1 million token context window, which can hold years of you at once. And I don't see that many people doing this.

So maybe you don't have old notes, but you definitely have emails. You definitely have text messages and ask it to find patterns in your decisions that you can't see because you're in the inside. And then just set up routines that remind you of things, right, and make you a better person. That's something that I think we're going to see more of. Number five. So fable is really good at noticing what's missing because it explores before it answers. So you can actually point that at anything, like your calendar or your codebase or your marketing funnel. So you can say, here's my analytics, my content calendar, and my offer. Don't tell me what's wrong with what exists. Spend your time on what's absent. What is every successful company in my space doing that appears nowhere in my business? What customer am I not even trying to reach? What revenue line should exist and doesn't? Ranked by money left on the table. So no-brainer on something that I started doing this morning.

have gotten some really interesting insights. Negotiation. You know, if you're in business, you're going to be doing negotiations. And a lot of people actually go to like perplexity or chat GPT and they'll say like, hey, I'm going to respond to this email. How would you respond? Now with Fable 5, you can actually actually have like basically a pro negotiator at the table. Here's what you say.

I'm negotiating a deal with this counterparty. Here's what I know about them and what I want. Become them. Their incentives. Their alternatives. Their pressure. We negotiate in rounds. You respond the way they actually would, including going quiet or getting aggressive.

After each round, break character and tell me what I just gave away for free. Don't let me win because I'm the user. So just easy way to just win more deals, get more. of them and here's the prompt. We're going to do two more use cases and we're going to get into startup ideas. So you can get a 80-page second opinion with Fable 5. So because Fable 5 actually reads tables now, footnotes, charts inside documents, it can see it, that's where the bodies are buried, right?

Leases, term sheets, insurance policies, vendor contracts, People are seeing incredible, I think it was Aaron Levy from Box, actually said that just in terms of legal, you know, they've been seeing just incredible outputs from Fable 5. So here's the prompt, here's the full contract, every exhibit included, read all of it, especially the tables and schedules, three lists. What cost me money that isn't obvious? what I'd get in 18 months, and what's missing that should be here to protect me. Then tell me the three changes you'd request and exactly how to phrase the ask. Flag anything where I should pay a real lawyer to look. Now, I'm not suggesting that you obviously just used AI to do your legal work.

But let's be real. Lawyers cost a lot of money. And if you're in a startup, every dollar counts. And some things, you know, I'm just speaking for myself. Some things I'm like, I don't need to send this to a lawyer. And some things, I just want to know where the flags are. So no matter if you're working with a lawyer or not, you know, do your own research. I'm not saying that you shouldn't work with a real lawyer.

I'm saying that even if you're working with a real lawyer, you should put it through Fable 5 and say, what are the flags? And here's the prompt that you're going to need for that. Boom. Make it build its own tools. This is the most meta one. It's the one that actually separates people who get it. So after you've used Fable 5 for a few weeks, just ask it to study the work and automate itself.

Here's what I mean. You prompt it. You say, look back at everything I've asked you to do this month. Find the request that repeat. Build yourself tools and reusable instructions for each one. So next time it's one sentence instead of 10. Then tell me, Based on what I keep asking, what should I be delegating to you that I'm still doing by hand like an animal?

And you'll be shocked. I've even done this just over the last 24 hours, what you've been able to automate and what you're able to, and what kind of software you're able to create here. All right. I'm going to go over a few startup ideas that are only possible now with Fable 5. I have five. I'll include the full list and the full prompts in the description. I'm only going to have time to go over three. So we're going to go over my three favorite ones. The first is a synthetic focus group firm. So D to C brands will spend like $50,000 on media before learning and ad doesn't work. So remember we talked about the copy tournament? Well, you run the tournament as a business. 50 ad variance. Judge over. by panels built from the brand's actual customer reviews, and then you charge something like $3,000 per launch, and maybe that costs you $100 of tokens. And the brand's media buyer becomes

your repeat customer. So this is only possible now because subagents can hold distinct personas and argue instead of a model agreeing with itself. Okay, here's what I mean. Here's the prompt. And this is, by the way, this is just going to be helpful for anyone, even if they don't want to use the startup idea, but if they're doing any paid ads themselves, this is going to be helpful, super helpful. So you give it 500 customer reviews. Here are 500 customer reviews for athletic greens

and their 10 best ads. Build six customer personas from actual review language, not demographics. Then take these 50 new ad concepts and have every persona score every ad. Would they stop scrolling? Would they click? What's their objection?

Output, a kill list, a launch list. And the one insight about this customer that the brand doesn't know. So someone in the comment section, I can hear them now. They're going to say, well, why wouldn't the DTC brands just do that themselves? Like if Fable 5 exists, you know, they're just going to go and do it themselves. They're not going to need your service. Well, that's where you're wrong. Yes. Well, you're right and you're wrong. You're right in the sense that some of them are going to do themselves, but most of them are not. Most of them might be scrolling on Instagram and they see an ad for this. And you hit them at the right time and you earn their trust. And it works. And they just get into the flow and you're in the mix. And you don't need that many clients to make a multi-million dollar business from it. So yes, you're a

definitely right in the sense of now anyone has these tools, people are going to use it, but the reality is not everyone's going to use it, and that's your opportunity. And if you can build a funnel around it, there's a business there. Another idea, custom software in 48 hours. Maybe you call this idea like 48 hours software.com. Every med spa contractor and clinic runs on spreadsheets, group tax, because custom software costs like $80,000 a year. So you sell them internal tools for $5,000 flat and you build it in two days. The interview prompt is your sales call. You run it with the owner on a Zoom.

The spec writes itself while they talk. Fable builds it after. Only possible now because it scopes vague messes itself. And scoping was always the expensive part. So I saw an X and it went viral that there was someone who was recording customer calls and they built like the customer maybe was complaining or something and they built an app while the customer was complaining. Let's get into the prop. You're about to interview a med spa owner about the most annoying repeated process in their business.

One question at a time, 12 question max. In plain English, no tech words. Dig until you can describe their workflow better than they. can. Then write the spec for the smallest tool that kills the annoyance, build it, and write a one-page guide their least technical employee can follow. Now you're creating internal software that they would have cost $80,000. Now they're getting it for $5,000. You're probably spending a couple hundred dollars max on tokens if you're token maxing. So the margins are incredible. And you can reinvest that money into the marketing to actually get people into this

funnel. I want to give you one more startup idea. So it's a contract refund firm. So midsize companies sit on hundreds of vendor contracts, but nobody has read those since signing. Auto renewals, unrenews seats, prices that ratchet 8% a year. You read all of them and then you get a paid 25% of the savings found. So it's a free product unless you find money for them. It's only possible now with Fable because it reads the tables and schedules insides hundreds of PDS without dying, which is exactly where the money hides. So here's the prompt. Here are all of our vendor contracts and 12 months of invoices. Cross-reference them. Find auto renewals in the next 90 days. Invoices that don't match contracted rates, seats we pay for.

but don't use. And the price escalators we can challenge. Rank by dollars. For the top 10, draft the exact email to the vendor, including the leverage we have and the number to ask. And I've got two more bonus startup ideas right here. So there you have it. Fable, five, the most powerful model to ever come out. And we're in this honeymoon phase where we you know, it's included in a lot of our plans that ends soon.

But I think it's important for right now just to get really, really good at what are the prompts, what are the use cases, what are the opportunities now with Fable 5 so that, you know, once it's API token based, we know what to use it for. So that's why I made this video. I made this, that's why I made this episode. This episode was designed to get your creative juices flowing around. What is now possible in the world of Fable 5? Because so much is possible.

And I wanted to narrow in on some ideas that are just so high value that to make sure you're not using Fable 5 around. Now you'll know how to create copy better than anyone else. Now you know how to create landing pages better than anyone else. Now you know how to, you know, different startup ideas going deep into data and how you can retool that, those ideas and how you can use Fable 5 to create those products. So hope this has been helpful. That's all I'm trying to do. Add value to your life. If you enjoyed this episode, please let me know in the comments. I read every single one. And all I got to say is I'm rooting for you and I can't wait to see what you build. I'll see you on the next one. Subscribe, like, and comment for more.