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

By the Income Reality Check Editorial Team · Updated 2026-07-26 · 7 min read

Mark Tilbury’s 7 ‘lazy’ AI money methods: what the effort scores skip

Verdict: Half-true — works only if you do the unspoken work. The tools cost $15–$50 a month and do exactly what he shows; the money still comes from selling, and none of the scores measure that.

Mark Tilbury’s “Laziest Ways to Make Money Online with AI (For Beginners)” has pulled in more than 120,000 views by promising something specific: seven side hustles where “AI can now do 90% of the work for you.” He scores each one on learning curve, how much the AI handles, startup cost, and competition, then ranks them from most crowded to least. Is any of it real? Mostly yes — and that’s exactly why the framing is slippery.

What the video actually claims

The through-line is a number: “AI does” 70% to 95% of each hustle, so your job shrinks to a sliver. Method one is AI clipping — chopping big creators’ podcasts into shorts with a tool like Opus Clips and getting paid through a marketplace such as Content Rewards. Tilbury shows a campaign offering “$1 for every 1,000 views,” so a 100,000-view clip pays $100. He rates the AI’s share at 90% and startup cost at “$15 a month.”

From there the ladder climbs in effort but shrinks in competition. AI ghostwriting (75% AI, $20–$40/month) using Claude Projects to mimic a client’s voice. AI landing pages (80%) with Wix or Hostinger. An “AI-built Shopify store” (95%) using a free store builder plus AutoDS — a segment he openly says AutoDS sponsored. Then AI phone-answering agents for local businesses (85%, “low” competition), AI avatars built in Higgsfield (70%), and no-code app building in Lovable or Emergent (80%, “very low” competition).

To his credit, Tilbury doesn’t pretend any of this is free money. He flags that clipping competition is “very high,” that “most people give up within a week,” and that landing your first AI-agent client “takes persistence.” The disclosure of the AutoDS sponsorship is the kind of thing the UK’s Advertising Standards Authority expects — its guidance says paid content should carry a clear label so viewers recognize an ad “without having to click or otherwise interact with it” (asa.org.uk). So this isn’t a scam video. It’s an accurate parts list with a misleading assembly time.

What the method actually requires

Look closely at the scores. “AI does 90%” is a claim about production — cutting the clip, writing the post, generating the store. Not one of these hustles pays you for production. They pay you for distribution and sales, and that’s the 10% the scores quietly shrink.

Take clipping, the “laziest” method. The AI genuinely edits and captions, but you’re paid per verified view, and views don’t come from the tool — they come from posting daily across multiple accounts and learning which hooks land. Independent reporting on the clipping economy is blunt about the curve: beginners earn little for weeks because minimum-payout thresholds mean low-view clips pay nothing, and steady clippers land in the low hundreds to low thousands a month, not the headline $100-a-clip. The $1-per-1,000-views rate is real. Hitting it reliably is a content job, not a lazy one.

The Shopify method is where “95% AI” strains hardest. AutoDS can pick products and fulfill orders, but “sending people to your store” — Tilbury’s entire description of your 5% — is the whole business. Dropshipping runs on thin margins (typically 15%–20% net), and you buy every customer with ads. Shopify’s own data puts customer acquisition cost at around $129 in fashion and accessories — the exact niche Tilbury picks in the demo — and up to $377 in electronics (shopify.com). If your product profit per order is less than that, you lose money on every sale. The “99 cents to get started” figure covers the tool trial. It does not cover the ad budget that decides whether the store earns a cent.

Here’s the pattern across all seven:

Who actually wins this game?

The same people who win most online-income games. Operators with a paid-traffic budget win the Shopify and landing-page methods, because ads — not the AI store builder — are the engine. People with existing sales chops win the AI-agent hustle; Tilbury rates its competition “low” precisely because it requires walking into dentists’ offices and closing a monthly contract, which most viewers won’t do. And anyone who already understands a niche wins at ghostwriting, where the paid skill is the “taste” Tilbury correctly calls the missing 25% — knowing what a finance founder would actually post, not typing a prompt.

Notice the tell in his own scoring: the methods he rates “low competition” are the ones that demand months of unglamorous B2B selling or audience-building. Competition is low there because the work is high. That’s not a loophole. That’s the market pricing effort correctly.

What you’d realistically earn

For a beginner putting in consistent hours, most of these land in the same unsexy range: roughly $0 for the first month or two while you learn distribution, then a plausible few hundred to a couple thousand dollars a month if you stick with one method past the point where “most people give up.” Clipping data supports that arc — real supplemental income, unstable, saturating fast. Ghostwriting is the clearest earner because clients pay retainers: entry-level rates run about $500–$1,500 a month per client, so two or three clients is a genuine part-time income. But those retainers go to people who can sell and edit, not to whoever downloaded Claude first.

None of that matches “laziest way to make money.” It matches “normal way to make money, with cheaper production tools.”

The gap between a real peak number and a typical one is exactly what regulators police. In 2024 the FTC ordered gig-work company Arise Virtual Solutions to pay $7 million after finding that 99.9% of people who joined its platform earned less than $18 an hour, despite the marketing (ftc.gov). The same agency ran a 2024 crackdown, “Operation AI Comply,” specifically targeting businesses that overstate what AI tools can do for buyers (ftc.gov). Tilbury is a UK creator ranking tools, not selling a business opportunity, so he’s not in that crosshair — but the underlying lesson is the same: a real top-end number tells you almost nothing about the median. If you want the broader version of this reasoning, see our breakdown of 5 proven ways to make money with AI and the wider survey in 5 ways to make money in 2026 with AI.

Who this is (and isn’t) for

This video is useful if you already accept that you’re starting a small business, not flipping a switch. You’ll want 8–15 focused hours a week, a little tolerance for months of near-zero returns, and — for the Shopify and landing-page routes — either an ad budget or genuine sales ability. The ghostwriting and AI-agent methods reward people who can talk to clients and don’t mind rejection. It’s a poor fit if you took “laziest” literally, have no capital for ads, and expect the AI to find customers for you. The tools produce; they don’t sell.

What to remember

Tilbury’s parts list is honest, cheap, and current — the AI really does cut the clip, write the post, and build the store. The scores are where it tilts, because “AI does 90%” measures the easy 90% and buries the hard 10% that actually gets you paid: distribution, ads, and client acquisition. Treat the video as a menu of legitimate tools, not a ranking of how little work you’ll do.