1
1 Comment

I spent 3 months building an AI girlfriend app and made $0. Then built a SaaS in 2 days that got its first paying customer.

Figured this might help someone else avoid the same mistake.

Earlier this year I went all in on the AI girlfriend trend. Chatbot first, then added image generation. Seemed like easy money - huge demand, people clearly paying for this stuff, how hard could it be?

3 months later I had:

  • 500 signups (paid ads on Traffic Junky)
  • 0 paying customers
  • A mess of problems I hadn't anticipated

The image generation was shit. Tried multiple models, none of them produced anything good enough. But that wasn't even the real problem.

The real problem: nobody wants to work with you in the adult industry.

Payment processors? Stripe, PayPal - instant no. The alternatives are sketchy, expensive, or both. Finding affiliates to promote it? Everyone ghosts you when they hear what it is. Even basic infrastructure stuff becomes 10x harder when you're in that space.

And the competition is insane. You're up against teams with actual budgets and years of head start.

I finally quit when I realized I was spending more time fighting the ecosystem than building the product.

The pivot

Took a week off, then asked myself a different question: what's a big, validated market where people actually pay, and I'm not fighting uphill just to exist?

Landed on B2B SaaS. Boring compared to AI girlfriends, but:

  • Payment processing just works
  • People expect to pay for business tools
  • I can talk about what I'm building without clearing my browser history

Built my current SaaS in 2 days using Claude. Not 2 months. 2 days.

It's been live for 2 weeks. I already have my first paying customer.

What I learned

The "hot trend" market isn't always the smart market. I got seduced by demand without thinking about everything else - payment infrastructure, reputation, competition, margins.

Sometimes boring is better. B2B SaaS isn't sexy but no one's blocking my Stripe account.

Speed matters more than perfection. My AI girlfriend app had 3 months of "polish" that got me nowhere. The new thing was ugly on day one but it was live and making money in 2 weeks.

If you're chasing a trend right now, ask yourself: even if people want this, can I actually build a business around it? Who will process payments? Who will promote it? Who's already dominating?

on January 28, 2026
  1. 1

    I’ve tried a bunch of AI companion apps, but the one that actually felt natural for me was at https://rubychat.app/ai-girlfriend because I could tweak the character’s vibe and set up little scenarios that made chats way more fun. I liked that my convos stayed private and I never got ghosted, which was a nice change of pace

Trending on Indie Hackers
AI runs 70% of my distribution. The exact stack. User Avatar 181 comments I'm a solo founder. It took me 9 months and at least 3 stack rewrites to ship my SaaS. User Avatar 145 comments I used $30,983 of AI tokens last month in Claude code on $200/mo plan User Avatar 52 comments We could see our AI bill, but not explain it — so I built AiKey User Avatar 25 comments AI coding should not turn software development into a black box User Avatar 24 comments my reddit post got 600K+ views. here's exactly what i did User Avatar 23 comments