Lovable Vibe Coding App Launches on iOS and Android

The AI development startup Lovable has officially launched its vibe coding app on iOS and Android, bringing no-code AI app building to mobile devices. The release arrives shortly after Apple enforced stricter guidelines on apps that download new code or alter functionality during runtime.

According to TechCrunch, the mobile app allows users to create working websites or web apps through voice or text prompts. The pitch is straightforward: capture ideas as they pop into your head, then let Lovable's AI agent run autonomously to build a prototype.

Here's where it gets interesting. Apple didn't ban vibe coding apps outright. The company blocked updates to popular tools like Replit and Vibecode for violating developer guidelines. The core issue: apps that download new code present security risks that Apple's App Review team cannot properly vet during approval. (It's a reasonable concern, honestly.)

Lovable adapted by shifting app previews to web browsers. This compliance move allowed the company to launch without the delays competitors faced. The app now focuses on building web apps rather than native iOS apps, aligning with Apple's updated policies.

iPhone in Canada reports additional features in the mobile release. The app includes a Visual Edits mode where users tap on elements and tell the AI to change them—shifting layouts, adjusting color schemes, or adding buttons without seeing a single line of code. For developers who want more control, the app supports two-way sync with GitHub, meaning the generated code is real and can be exported or refined by professional teams later.

The workflow is designed for speed. Users can start a project by typing a description or uploading a screenshot of a design they like. From there, the AI builds a working prototype in real-time. The app syncs across devices, allowing seamless transitions between computer and phone. Notifications alert users when a build is ready for review.

This approach reflects a broader trend in AI-powered development. The term "vibe coding" gained popularity after being coined by AI researcher Andrej Karpathy. It describes development where you don't write line-by-line code. Instead, you describe the functionality in plain English. The AI handles the heavy lifting, generating the frontend, setting up databases, and managing authentication.

Early testers have noted that while the AI is incredibly fast, complex logic still requires back-and-forth refinement to get exactly right. The friction isn't in typing code—it's in articulating what you actually want. (Anyone who's tried to explain a UI design to a developer knows this pain.)

The no-code AI app builder market is growing rapidly. Lovable's launch intensifies competition with platforms like Bubble and Adalo. By offering a mobile-first experience, Lovable differentiates itself in a crowded space. Industry analysts predict that mobile vibe coding will attract millions of new users, particularly in emerging markets where smartphone penetration is high.

Apple also temporarily removed the vibe-coding app Anything from the App Store for similar reasons. The app returned after making changes earlier this month. This pattern suggests the platform isn't hostile to the category—it just wants guardrails.

Lovable's compliance with Apple's rules sets a precedent for other developers. The company demonstrated that innovation can coexist with platform policies. This balance is crucial for long-term viability, as app store regulations continue to evolve.

The physical experience matters here. Holding a phone and speaking a prompt feels different from typing in a code editor. The tactile feedback of tapping through a generated UI, waiting for a notification that your build is ready—these are real interactions, not just abstract concepts. Whether that translates to actual productivity remains the real question.

Whether users actually pay for it remains the real question.

Artūras Malašauskas is an AI Systems Integrator with 20+ years of production-grade web engineering experience. He has designed, shipped, and scaled enterprise Python/PHP systems for logistics, SaaS, and public-sector clients. For the past year, he has focused exclusively on AI integrations: deploying open-source LLMs, building generative media pipelines (image, audio, video), and engineering multi-agent workflows for real production environments. His standard: reproducibility, security, cost-efficient inference—no vaporware. He documents and evaluates emerging AI tooling, separating verified capabilities from marketing noise. Technical editor at: muza-ai.eu, ai-verslas.lt, ai-naujinos.lt Connect on LinkedIn