How to Start a Side Hustle with AI (Even If You Have Zero Tech Skills)
I have a notebook. It's navy blue, spiral-bound, lives in my nightstand drawer. Inside it are approximately forty-seven business ideas I have written down between the hours of 11pm and 2am over the past six years.
Custom stationery. A meal prep service for busy moms. An online course about something. A photography mentorship. A candle line. (I have never made a candle in my life.)
Every single one of those ideas made complete sense at 1am. And every single one of them died the next morning when I opened my eyes, looked at my actual schedule, and thought: where would I even start?
Not the idea part. I've always had ideas. The other part. The writing-a-proposal part. The figuring-out-what-to-charge part. The sending-a-cold-email-to-a-stranger-without-wanting-to-crawl-under-my-desk part.
That's the part AI can help with. Not the dreaming. The doing.
First: Figure Out What You Actually Want to Do
Here's what I see happen all the time. Someone decides they want to start a side hustle, so they Google "best side hustles 2026" and end up reading a list that includes "start a dropshipping business" and "learn to code." Then they close the laptop and go watch TV because none of that sounds like them.
The better move is to start with what you already know how to do. You have skills. Real ones. The problem is you've been doing them for free or for an employer for so long that you've forgotten they're worth money.
Open ChatGPT (chatgpt.com, free) or Claude (claude.ai, also free) and try this:
The key part of that prompt is "based on what I already know." Without it, you'll get a generic list that has nothing to do with your actual life. With it, you get ideas that make you think, "Oh. Yeah. I could actually do that."
When I ran this for myself, one of the suggestions was offering brand photography packages for small businesses. Which is literally what I ended up building. Sometimes you need a machine to point out the obvious thing you've been standing on top of.
Name the Thing
I spent three weeks trying to name Soft Tech. Three weeks. I had a Google Doc with over sixty names in it and I hated all of them. Some were too cute. Some were too corporate. One of them was "Digital Divas" and I still lose sleep over the fact that I considered it for more than four seconds.
AI won't hand you the perfect name. But it will generate fifty options in thirty seconds, and somewhere in that list, something will click. Or two of them will half-click and you'll mash them together and get something that works.
That last line matters. "Something I'd be proud to put on a business card" filters out a lot of noise. Without it, you get names that sound like discount mattress stores.
Write Emails That Don't Sound Desperate
OK so here's where most people stall out. You have the idea. You maybe even have a name. But now you need to actually tell someone about it. You need to reach out to a potential client, or respond to someone who asked about your services, or follow up with that person you met at a networking event who said "we should talk."
And you sit there staring at a blank email for twenty minutes because everything you type sounds either too salesy or too apologetic.
I used to start every client email with "I hope this finds you well" and end it with "no pressure at all" and wonder why people didn't take me seriously. The answer was that I was pre-apologizing for my own existence.
Here's the prompt I use now for outreach emails:
Read it out loud before you send it. If any sentence makes you cringe, change that sentence. You're looking for something that sounds like you on a good day, when you're confident and not overthinking it.
Write a Proposal Without Spiraling
The first time someone asked me to send a proposal, I panicked. What goes in a proposal? How long should it be? Do I need a cover page? Is there a font that says "I'm professional but also fun"? (There isn't. I checked.)
A good proposal does four things: it shows you understood what the client needs, it explains what you'll do, it tells them what it costs, and it gives them a timeline. That's it. You don't need a 12-page PDF with your mission statement and a mood board.
Copy that into a Google Doc, add your name and contact info at the top, and you have a proposal. The whole thing takes maybe ten minutes. The first time I did this, I sat there looking at the finished document thinking, "That looks like something a real business person would send." Which was the whole problem. I was a real business person. I had been the whole time. I was the one who didn't believe it.
The Part Nobody Talks About
Here's what I want you to hear if nothing else from this post lands.
The gap between "I have a skill" and "I have a business" is not a knowledge gap. It's a confidence gap. You already know how to do the thing. You've probably been doing the thing for years. What stops you is the business-y stuff that surrounds the thing: the emails, the pricing, the proposals, the "what do I even call myself" spiral that eats up your entire Saturday.
AI doesn't give you talent. You already have that. It gives you a first draft of all the stuff that was keeping you stuck. The email you couldn't write. The proposal you didn't know how to format. The business name you couldn't land on.
You're not starting from zero. You're starting from a lifetime of skills and a blank page. And now the blank page part is handled.
Open that notebook in your nightstand. Pick the idea that still has a little heat to it. The one you keep coming back to. Run the first prompt. See what comes back.
That's your move.