You can make an app for that

AI coding tools such as Claude Code, OpenAI’s Codex, and others are now capable of helping individuals build functional, personalized software with minimal technical expertise, marking a significant shift in software creation and use.

Recent updates to AI models have made it possible for users to develop their own apps using simple descriptions and prompts. These tools, once primarily used by professional developers, are now accessible to the general public, enabling the creation of bespoke software for tasks like managing family budgets, custom productivity tools, or business automation. Experts like Andrej Karpathy have described this phenomenon as ‘vibe coding,’ emphasizing its intuitive, user-friendly nature. However, these personal apps are often hacky, lack formal support, and are not guaranteed to be secure or thoroughly tested. Despite these limitations, the ability to craft tailored solutions is empowering many users to replace or augment existing software with their own creations.

Why It Matters

This development matters because it democratizes software development, allowing individuals to address their unique needs directly without relying on commercial software vendors. It could lead to a shift in how people interact with technology, fostering a new wave of personalized digital tools that are more aligned with individual workflows and preferences. While not replacing professional software entirely, this trend highlights the growing importance of AI as a tool for empowerment and customization in everyday life.

Background

The concept of personal software has been around informally, as exemplified by Robin Sloan’s 2020 blog post about building simple, home-made apps. The recent AI advances have dramatically lowered the barriers, transforming hobbyist and non-technical users into software creators. This shift follows the broader AI boom of late 2025, when models like Claude Code became unexpectedly capable of generating and fixing code based on natural language prompts, leading to the rise of ‘vibe coding.’ Prior to this, most software customization was limited to professional developers or through limited automation tools like IFTTT or Apple Shortcuts. Now, AI is enabling a new paradigm of user-driven software creation.

“Suddenly, all you needed was $20 a month and a half-formed idea, and an AI model could build you functional software.”

— Andreja Karpathy

“There will be no sudden redesign, no flood of ads, no pivot to chase a userbase inscrutable to us. It might go away at some point, but that will be our decision.”

— Robin Sloan

What Remains Unclear

It remains unclear how widespread or sustainable this trend will become, especially given the limitations of personal software such as security, support, and testing. The long-term impact on professional software markets and enterprise solutions is also uncertain.

What’s Next

Developers and users will likely see continued improvements in AI coding tools, making personal app creation easier and more reliable. Regulatory and security considerations may influence adoption, and we can expect further experimentation with AI-driven software customization in various domains.

Key Questions

Can anyone create their own app with AI tools?

Yes, recent AI coding tools are designed to be accessible to non-professionals, enabling users to build simple, functional apps based on natural language prompts.

Are these AI-generated apps secure and reliable?

Most personal apps created with current AI tools are hacky and lack formal testing or security guarantees. They are suitable for personal use but not for critical applications.

Will AI replace professional software developers?

While AI makes personal software creation easier, it is unlikely to replace professional developers entirely, especially for complex, secure, and enterprise-grade solutions.

How does this trend affect the software industry?

This trend could lead to more user-driven customization and innovation, potentially reducing reliance on commercial software for niche or personal needs, but it is unlikely to disrupt large-scale enterprise software markets significantly.