There’s no denying that you can use artificial intelligence to make money.
However, it’s important to differentiate between legitimate business opportunities and temporary arbitrage plays.
Way too many people are pursuing the latter without realizing that those income streams might vanish overnight.
That’s why today we want to share five ways to make money with AI that you can leverage to build a real, sustainable business.
This is a long article so feel free to skip around using the table of contents. We do recommend reading it in its entirety, though!
What You Need to Know About AI
What is AI?
“Artificial intelligence” – AI for short – is an umbrella term.
It encompasses technologies that enable machines to perform cognitive tasks that are associated with human intelligence.
Examples of such cognitive tasks include decision-making, problem solving, learning from previous experience, data analysis, content creation, and visual perception.
What is Generative AI?
Nowadays, when people talk about artificial intelligence, they are typically talking about generative AI.
Generative AI is a type of artificial intelligence that can be used to generate novel data based on user input.
Do you know how there are all these AI apps that you can use to create text, images, and videos? That’s generative AI!
What are Large Language Models (LLMs)?
Large language models are a type of generative AI that can understand the text in user queries and generate novel, relevant text in response.
They are trained on vast quantities of textual data which enables them to mimic human speech and writing patterns.
This is the technology behind chatbots like ChatGPT that can have unstructured conversations with their users.
What are the Most Popular Large Language Models?
At the time of writing, the three most popular large language models are:
These chatbots appear to be quite similar so we recommend trying out all of them to see which one you like the best.
You can experiment by giving them the exact same prompts and then comparing their outputs.
What is the Black Box Problem?
This might sound weird but…
It seems that computer scientists at OpenAI, Anthropic, and Google don’t know how their own large language models work.
In fact, to the best of our knowledge, no one does. Scientists cannot see what is happening inside LLMs. All they can observe is the data that goes in and the data that comes out.
This is known as “the black box problem” and it’s arguably the greatest bottleneck in artificial intelligence research. Unfortunately, it’s unlikely that it’s going to be solved anytime soon.
What are “Hallucinations”?
Since scientists don’t know what is happening inside large language models, they cannot ensure consistently accurate output.
The issue is that sometimes LLMs just make things up. This is known as “hallucinations”.
So far, no one has been able to figure out how to get rid of them. In all likelihood, large language models will continue hallucinating for the foreseeable future.
In practice, this means that you should never blindly trust their output because there is no guarantee that it’s accurate.
For example, if you use ChatGPT to write social media posts, make sure to manually fact-check everything before you publish them.
This problem isn’t limited to LLMs, though. Generative AI models designed to produce images also hallucinate.
Say, one common tell that a photorealistic image generated with AI is too many fingers on people’s hands. Seems obvious but it’s easy to miss if you aren’t paying attention.
If you want to use AI to produce images, make sure to closely examine them before publishing.
Finally, generative AI models designed to produce video content aren’t immune to hallucinations either.
For example, here’s a screenshot from a video that was created with OpenAI’s Sora:
Here’s the prompt that was used to generate this video:
“The Glenfinnan Viaduct is a historic railway bridge in Scotland, UK, that crosses over the west highland line between the towns of Mallaig and Fort William”.
The problem is that Glenfinnan Viaduct only has one railway track. There are more inaccuracies in the video but this is the most noticeable one!
Just like with AI-generated images, if you want to use AI-generated videos, you need to examine them closely before publishing. Does everything look right?
What is Prompt Engineering?
The quality of your prompts is arguably the main factor that determines the quality of generative AI output.
Prompt engineering is the practice of crafting effective prompts that produce the type of output that you want.
Everyone who uses generative AI on a regular basis eventually develops their own prompt engineering style.
However, if you are new to this, we recommend starting with Jeff Su’s prompt formula which consists of these six components:
- Task: mandatory
- Context: important
- Exemplar: important
- Persona: nice-to-have
- Format: nice-to-have
- Tone: nice-to-have
As you can see, the components are listed in descending order, the “task” component being the most important and the “tone” component being the least important.
Here’s how Jeff Su explains it:
Where Can You Learn More About AI?
If you are serious about harnessing the power of artificial intelligence to make money, consider taking Google’s “AI Essentials” course that is available on Coursera.
It consists of five modules:
- Introduction to AI (1 hour)
- Maximize productivity with AI tools (2 hours)
- Discover the art of prompting (2 hours)
- Use AI responsibly (1 hour)
- Stay ahead of the AI curve (2 hours)
If you want to get a feel for what this course is like, you can watch the first module for free on YouTube:
“AI Essentials” focuses on teaching people how to effectively use Google’s large language model Gemini.
However, a lot of what you’ll learn will also apply to OpenAI’s ChatGPT, Anthropic’s Claude, and other LLMs!
Legitimate Business Opportunities vs. Temporary Arbitrage Plays
How to Recognize Legitimate Business Opportunities
Understanding the Lindy effect can be helpful here.
It states that the future life expectancy of some non-perishable things is proportional to their current age. To put it simply, the longer something has been around, the more likely it is to stay around.
That’s why we believe that the safest bet is to start a business that provides a service that already has proven demand and then use artificial intelligence to increase your company’s output.
This approach is Lindy because it enables you to capitalize on business opportunities that have been around for quite a while and are therefore unlikely to disappear anytime soon.
Meanwhile, pursuing novel business opportunities that have popped up since the release of ChatGPT in 2022 is not Lindy. Why?
Because something that has only been around for a few years is unlikely to still be around a decade from now. Who knows, maybe it will. But we wouldn’t bet on it!
What is the Problem With Temporary Arbitrage Plays?
The problem with temporary arbitrage plays is that they tend to go against the interests of the platforms that they are leveraging.
For example, Amazon has become a go-to place for buying books, both paper and digital. It’s in its best interest to keep its customers happy.
On top of that, the company also needs to keep the book authors happy, including the self-published ones who are using its Kindle Direct Publishing service.
Once generative AI became widely available, Amazon got flooded with low-quality books produced with it.
Needless to say, customers who accidentally bought such books were not happy about it.
Meanwhile, authors were not amused by the fact that their books were being crowded out by AI slop.
There have also been several high-profile incidents when immediately after the release of a long-anticipated book a bunch of AI copycats popped up trying to steal its sales.
Prominent tech journalist Kara Swisher event sent an email to Amazon CEO Andy Jessy asking “’What the f***?’ You’re costing me money” when this happened to her.
Do you think Amazon is going to allow this AI mess to continue when it’s making both the customers and the authors angry, not to mention all the negative publicity the company has received due to this?
Questionable ethics aside, some people have made a lot of money with low-quality AI-generated books, but it’s probably just a matter of time until Amazon develops software for automatically detecting and removing AI slop.
This is the lifecycle of temporary arbitrage plays: hustlers see an opportunity and jump on it, and some of them even make good money with it, but eventually, the platform cracks down on them and everyone gets banned.
Resist the Siren Song of Get-Rich-Quick Schemes!
If you go on YouTube and search “how to make money with AI”, you are going to see a bunch of videos promoting temporary arbitrage plays as a way to get rich quick.
It can be very tempting to pursue these opportunities when you see all these content creators talking about how much money they are supposedly making.
However, as we explained in the previous section, income sources like that might be here today and then be gone tomorrow with no warning.
Some people waste years of their lives chasing quick money without building anything lasting. Don’t fall into this trap because you are going to regret it.
Instead, resist the siren song of get-rich-quick schemes and focus on building a legitimate business. The future you is going to thank you for that!
Best Ways to Make Money With AI
Okay, so with all that out of the way, let’s get to the practical stuff. How can you make money with AI?
#1: Social Media Marketing Service
Social media marketing can be a great way to attract customers.
However, many small businesses, especially local ones, are underutilizing this marketing strategy.
Typically, the owners of these businesses know that they should be posting regularly on social media, they just never get around to it. But they might be willing to pay you to do it!
Consider using:
- ChatGPT, Claude, or Gemini to brainstorm ideas
- Jasper to turn those ideas into social media posts
- Adobe’s Firefly to generate social media images
- Buffer or MeetEdgar to schedule your posts
- Hootsuite to manage everything in one place
This software stack makes it easy to consistently produce high-quality content. And that’s what success on social media is all about!
#2: Podcast Editing Service
Everyone wants to have a podcast these days but no one wants to spend countless hours editing raw audio files. So why not offer a podcast editing service?
With the AI tool Descript, you can edit audio by editing its transcript so it’s much less tedious than it used to be. Basically, if you know how to edit text documents in Google Docs, you will be able to edit podcasts with this app.
It also has a bunch of neat podcast-specific features that will surely come in handy during the editing process!
#3: Video Editing Service
Video is becoming the dominant medium of the Internet, which means that video editing skills are more prized than ever.
There’s also a growing demand for producing short-form content where the editing needs to be really tight in order to provide value to the viewer.
Video editing used to require mastering complex software like Adobe Premiere Pro and Final Cut Pro. And if you want to be good at it, you still need to learn how to use these tools.
But you can get started with the aforementioned AI app Descript which allows you to edit videos by editing their transcripts.
Moreover, you can run long-form videos through OpusClip and its AI is going to automatically identify the best moments and turn them into shorts!
#4: Video Thumbnail Design Service
The thumbnail of a video is arguably the most important factor that determines how many views it’s going to get on YouTube.
With generative AI tools like Adobe’s Firefly, you don’t need to be a graphic designer to create attention-grabbing thumbnails that people can’t help but click.
There are a lot of YouTubers out there that have great content but their boring video thumbnails are sabotaging their growth.
So why not help them revamp their channels with amazing thumbnails?
#5: Website Design Service
Some businesses have websites that look so outdated that they can make you feel like you time-traveled back to the early 2000s.
It’s another one of those problems that business owners tend to be aware of but never get around to fixing.
If you are a tech-savvy person, you can offer them a complete website makeover where you set up hosting for them, redesign their website, and rewrite its copy.
Popular website builders like Squarespace and Wix now include AI editors so you don’t need to be a web designer to create beautiful custom websites.
As for the copy, we recommend reading our friend Jim Edward’s book “Copywriting Secrets” where he explains how to write copy that converts.
Once everything is done and the website is live, you can pitch the client a maintenance retainer where you will update their website for them.
This might be appealing to people who don’t want to deal with tech stuff!
How to Build an Online Service Business
Clearly Define Your Dream Clients
Your dream clients must:
- Need your service
- Be able to pay for your service
- Be willing to pay for your service
Your target audience will be the main determining factor when it comes to the revenue potential of your company. Go after established businesses if you want to maximize it!
Start by Acquiring Social Proof
Social proof is the evidence that other people trust you. In the business context, it primarily means customer reviews, testimonials, and case studies.
The more social proof you have, the easier it will be for you to get clients, which is why you should focus on acquiring it first. So how can you do that?
You’re probably not going to like the answer: the easiest way to get social proof is to do free work in exchange for testimonials.
For example, if you want to acquire social proof for your podcast editing business, identify podcasts that could use your help.
Then, edit their most recent episodes and send the hosts sound files. Tell them that you can edit 12 more episodes for free in exchange for a testimonial.
Start at the bottom of the hierarchy by reaching out to up-and-comers who probably need your service but may be hesitant to pay for it.
Remember, right now the focus is getting social proof, not making money!
See if You Can Turn Free Work Into Paid Work
Okay, so you completed the free work and have your testimonial. Now what?
Let that potential client know that they can hire you if they want to and name the lowest price possible.
If they are happy with your free work and can afford to pay you, they will probably end up hiring you.
And if they don’t want to hire you now for whatever reason, tell them to feel free to reach out to you if they ever need your service in the future!
Use Cold Outreach to Grow Your Client Base
Start reaching out to your dream clients via email and social media DMs.
Make sure to optimize every cold message by referencing something about that particular potential client.
Then, pitch them a clearly defined package of free work that they can use if they like, with no strings attached.
Follow up that pitch with your most powerful piece of social proof and then provide a link to your business website if they want to read more testimonials.
Finally, end the message with a call to action. Ask them if they are interested, invite them to schedule a Zoom call with you, and share your appointment booking calendar link.
If you don’t get a response, send them a follow-up message one week later and then continue following up once a week until you get a yes or no answer from them.
In your follow-ups, include a P.S. section similar to this one that the founder of LeadFuze Justin McGill used in his cold email campaign:
Get Good at What You Do Before Scaling Your Business
Entrepreneurs often attempt to scale their businesses too early, which leads to negative word-of-mouth.
Realistically speaking, if you just started providing a particular service, you are probably not that great at it.
If you offer to do free work for someone, they likely won’t be expecting much. That gives you an opportunity to learn.
However, once you start charging for your service, the stakes get higher. If people pay you and you disappoint them, they will be unhappy about it and might complain to others. And if enough clients do that, it can ruin your reputation.
That’s why it’s best to focus on getting good at what you do before you worry about scaling. Otherwise, you might end up accidentally scaling brand damage!
Scale Your Business With a Webinar Funnel
Once you are ready to scale your business, you can do that with a webinar funnel.
It should have three pages:
- The registration page is where you introduce your webinar and encourage potential clients to register for it.
- The registration confirmation page is where you reassure potential clients that they have successfully registered for your webinar.
- The webinar page where the potential clients can attend your webinar live or watch the replay.
As for the webinar itself, we recommend structuring it like this:
- A 45-minute presentation on a topic related to your service.
- A 15-minute pitch for your service where you explain what it’s all about, provide social proof, and invite the attendees to book a free consultation with you.
You can use paid advertising on platforms like Facebook, Instagram, and LinkedIn to drive traffic to your webinar registration page.
Although it’s possible to automate this funnel by using a pre-recorded webinar, we recommend doing it live once a week for at least a year.
That will help you learn what resonates with your target audience and enable you to optimize your webinar script for conversions!
Want to Learn How to Build Sales Funnels That CONVERT?
Our co-founder Russell Brunson used sales funnels to take ClickFunnels from zero to $100M+ in annual revenue in less than a decade.
He is now widely considered to be one of the top sales funnel experts in the world. Want to learn from him?
His best-selling book “DotCom Secrets” is the best place to start because it covers everything you need to know in order to build sales funnels that convert.
This book is available on Amazon where it has over 2,500 global ratings and a 4.7-star overall rating.
But you can also get it directly from us for free…
All we ask is that you pay for shipping!
So what are you waiting for? 🧐