We’ve all been there. You’re sitting around with friends, maybe after a few drinks, and someone says, “You know what would make a great app?” You’re excited, thinking about features, and maybe even checking to see if the domain name is taken. And then? Nothing happens. The dream dies right there because no one at the table really knows how to program. Learning Swift or hiring a developer is expensive and difficult, so the idea just ends up gathering dust.
Replit is essentially trying to kill that excuse forever
They just launched a new tool called Mobile Apps on Replit, and honestly it seems like we’re finally seeing the true promise of AI. ChatGPT doesn’t just write a broken Python script for you. It’s a tool where you can literally type in plain text, “Make me an app that tracks the top 10 companies by market cap” and the whole thing is set up. Interface, logic, everything. It even connects to Stripe so you can start charging people right away. They claim that you can go from a text prompt to a live app in the Apple App Store in just a few days. That’s wild.
It’s no wonder investors are spending money on it. Rumor has it that Replit is targeting a valuation of $9 billion in its next round of funding. This is a huge jump from where we were a few months ago. It reflects this huge explosion of what people call “vibe coding.” It’s a funny term, but it’s fitting – you don’t write strict instructions anymore; You just describe the “mood” or desired outcome, and the AI does the heavy lifting. We’ve seen it with tools like Anthropic’s Cursor and Claude, which also drive insane sales numbers.
But here’s where things get interesting – and a bit messy
First of all, Apple is still the bouncer in the club. You might be able to build an app in ten minutes, but you still have to get it through the App Store review team. Apple claims to review 90% of apps within 24 hours, but is notoriously concerned about privacy and quality. Just because an AI created it doesn’t mean Apple will approve it.
Then there’s the scary part: security. A cybersecurity firm called Tenzai recently investigated this and found that apps created by these “vibe coding” agents often come with vulnerabilities. AI focuses on making the code work, not necessarily on making it secure. It could be that basic things like brute force protection are forgotten, leaving your shiny new app open to hackers. And if you as a “developer” don’t really know how to read code, you wouldn’t notice the vulnerability until it’s too late.
So while traditional software stocks are taking a hit because everyone thinks AI will replace developers, the reality is a little more nuanced. We’re definitely entering a new era where building software is as easy as sending a text message, but we may also be flooding the world with apps that are perhaps a little… fragile. It’s an exciting time to be an idea generator, but perhaps a terrible time to be a security auditor.
