AI Side Hustles Half-true — works only if you do the unspoken work

By the Income Reality Check Editorial Team · Updated 2026-09-09 · 7 min read

Claude AI in 24 hours: the free $10k/month business Mr Reis doesn’t finish showing you

Verdict: Half-true — works only if you do the unspoken work. Claude Code really can build the app, the brand, and the sales page for free in an afternoon. Getting anyone to pay for it is the other 95% of the job, and the video ends before that starts.

Mr Reis opens his video “Easiest FREE Way To Make Money With Claude AI (Within 24 hours)” with a wall of success stories — Pieter Levels making six figures a month from a selfie app, a guy selling hummingbird sugar who cleared $5,000 in 21 days, a founder whose store-builder app passed $250,000 in profit. Then the pitch: with Claude Code and zero skills, zero followers, and no cold DMs, you can spin up a “boring business” that reaches $10,000 a month. Is that real? The building part mostly is. The earning part is a different story, and it’s the part he speeds past.

What the video actually claims

The method is straightforward, and to his credit Mr Reis shows every step on screen. You list things that annoy you or subscriptions you pay for. You paste that into Claude Code and ask for ten “boring business” ideas. Claude picks one — in the demo, an AI service that drafts replies to Google reviews for local shops at $199/month — and then builds a full plan around it: a brand name (Ravees), a logo prompt, a spreadsheet showing you need 51 clients to hit $10,000/month, a one-page sales site, and even a working desktop app that generates review replies.

The setup, he says, takes 30 minutes and costs $0: a free Gmail, a free logo from an image generator, free Netlify hosting, a free Stripe account. Promotion is a video ad made by the sponsor’s tool. The plan Claude produces lands the $10,000 target “around month six” at 8–10 new clients a month.

He’s careful with one line: “You don’t usually start making $10,000 a month, but these businesses have the potential to take you there.” Hold onto that sentence. It’s doing a lot of work.

What the method actually requires

Here’s the first thing the video glosses. Claude Code isn’t free. The version that builds apps and drives your browser lives on Anthropic’s paid plans — Claude Pro is $20/month (about $17 if paid annually), and the free tier doesn’t include Claude Code at all. Mr Reis admits this near the end and promises a future video on a “free” workaround. So the honest starting cost is $20/month, not zero.

The bigger gap is who does the selling. Watch the plan closely: “Find 50 local businesses on Google Maps with ignored reviews. Message 10 a day.” That’s cold outreach. It’s exactly the “sending DMs” and “starting an agency” work the intro promised you’d never have to do. The AI writes the replies; it does not find the customers, handle the “no thanks,” or chase the invoices.

Then there’s a technical problem the demo quietly sidesteps. Posting replies to Google reviews on a client’s behalf isn’t a free-for-all. Google’s own documentation is clear that a business controls its own profile and its own replies, and third-party review content is governed by Google’s rules, not yours (see Google Business Profile Help). The app in the video is a local desktop tool that drafts a reply for the owner to copy and paste — which is fine, but it’s a long way from the hands-off “it posts for you” impression.

Getting paid has a cost too. Stripe is free to open, but it takes a cut. On U.S. online card payments that’s 2.9% plus 30 cents per transaction, per NerdWallet’s Stripe breakdown. On a $199 subscription that’s about $6 a month, every month, per client. Small, but it’s not “$0.”

How hard is the money, really?

Let’s take the plan at its word: 51 clients at $199/month. That’s not a side project. That’s a real book of business you have to sell, onboard, and keep from canceling. And churn is the silent killer here. Independent micro-SaaS data pegs the median one-person software product at roughly $500/month in revenue, with only a small minority ever clearing $10,000/month — and healthy products still lose several percent of customers every month. Signing 51 clients isn’t the finish line. You have to keep re-signing the ones who quit, on top of adding new ones.

Now the math the video doesn’t run out loud. If you message 10 local businesses a day, five days a week, that’s about 200 pitches a month. Local B2B cold outreach converts in the low single digits at best. Land 2% and you’ve got four new clients that month — decent, but a long way from 8–10, and you haven’t accounted for the ones who churn out the back door. The “month six” timeline assumes near-flawless conversion and near-zero cancellation. Real numbers push that further out, if it happens at all.

Who actually wins this game

The people who make local-services SaaS or agencies work usually share one of three things: an existing audience they can pitch to for free, a budget to run paid ads (Mr Reis quietly folds “run ads to this content” into the plan — that’s real money), or prior sales experience so cold outreach doesn’t feel like walking into a wall. The founders name-dropped in the intro — Pieter Levels, the app builders with tens of thousands of users — had years of shipping and large followings before the win the video celebrates. That’s the context the highlight reel leaves out.

If you have none of those advantages, you’re not doomed. But you’re the person doing the grinding, unglamorous outreach for months while the AI handles the easy 5%.

What you’d realistically earn

Set expectations against the claim. Mr Reis frames $10,000/month as the destination, month six as the ETA. A realistic path for a beginner with no audience: roughly $0 for the first month or two while you learn to pitch, maybe one to three clients ($200–$600/month) by month three or four if outreach clicks, and — if you’re consistent and good at retention — a few thousand a month somewhere in year one. Plenty of people never get past their first two or three clients. That’s not a knock on Claude; it’s the base rate for cold-started service businesses.

The FTC has spent years going after exactly this genre of promise. In January 2025 it proposed new rules specifically to curb deceptive earnings claims from “business coaching and other schemes,” warning that inflated income figures “lure people looking for honest income into spending thousands” (FTC). Mr Reis isn’t selling a course, and his “potential to take you there” caveat is more honest than most. But a headline of “$10k/month in 24 hours” and a body that quietly means “$10k/month maybe by next year if you sell hard” is the exact gap regulators watch. (U.S. readers: those FTC rules apply to sellers in the U.S.; the underlying reality — income needs distribution — is universal.)

Who this is (and isn’t) for

This fits someone who has a few hours a week for outreach, is comfortable talking to local business owners, and treats the AI as a production assistant rather than a money printer. Twenty dollars a month and a willingness to hear “no” fifty times is the real entry fee. It does not fit someone expecting passive income, someone who won’t do sales, or someone who took “within 24 hours” literally. The build takes an afternoon. The business takes the rest of the year.

If you want a broader map of what these tools can and can’t do, our roundups on Claude AI side hustles that can pay a full-time income and proven ways to make money with AI are worth a read next to this one.

What to remember

The demo is genuinely impressive — Claude Code will plan, brand, and prototype a business faster than you can write the to-do list. That’s the real half of the pitch. The missing half is that a business nobody has heard of earns nothing, and closing that gap is cold outreach, ad spend, and retention over months. The AI builds the storefront in a day. You still have to go find the customers.