Vibe-Coding: The Future of Code or Just a 'Short-Term Con'?

In a world racing to use AI for anything and everything, Collins Dictionary has captured the mood of the era, announcing vibe-coding as its Word of the Year for 2025.

The term has taken the technical world by storm. It's widely popular for its ability to democratise coding, allowing people with little or no technical knowledge to create functional applications. But it's also deeply controversial, sparking debates about its questionable real-world effectiveness. 

Christened in February 2025 by OpenAI co-founder Andrej Karpathy, the word has since blown up and seeped into our everyday lexicon.

“There's a new kind of coding I call "vibe coding", where you fully give in to the vibes, embrace exponentials and forget that the code even exists,” said Andrej on X, sparking a new age of coding.

What is vibe coding?

Collins Dictionary describes vibe coding as "the use of AI prompted by natural language to assist with the writing of computer code".

Tools like Replit, Cursor, Google AI Studio and Microsoft GitHub Copilot enable the development of fully functional applications using prompts from users with varying levels of experience, or even none at all.

Andrej noted that vibe coding is only possible because LLMs are “getting so good”. 

“Sometimes the LLMs can't fix a bug, so I just work around it or ask for random changes until it goes away.

“It's not too bad for throwaway weekend projects, but still quite amusing. I'm building a project or web app, but it's not really coding - I just see stuff, say stuff, run stuff, and copy paste stuff and it mostly works.”

How much code is now written through vibe coding?

Sundar Pichai, Google's CEO, reveals that over 30% of new code developed at Google comes from AI. 

A recent Fastly research revealed that over half of the code released by a third of the senior software developers were AI generated.

Junior developers seem more reluctant to embrace the vibe, as only around 13% shipped code, more than half of which were developed by AI.

This gap can be explained by the time required to debug and fix mistakes generated by AI tools. 

Senior developers with over 10 years of experience, are more adept at catching mistakes and fixing bugs, making it a tool of convenience that equips them to code faster, while junior developers struggle to create real productivity gains from it. 

Dangers of vibe coding: Is it a 'short term con'?

Vibe coding has sparked debates from its origin on its effectiveness. 

Bug-prone code lacking complete architectural structure is difficult to debug and integrate into enterprise systems.

Raymond Kok, CEO at Mendix, a Siemens Company says that while vibe coding is fast and creative, it is deeply unreliable for enterprise use.

“Businesses must not get ahead of themselves and blindly follow the vibe coding trend without full consideration of how it will affect wider IT strategy. 

“Organisations must take time to build a comprehensive strategy fit for the long haul. 

“When setting out governance plans, they should keep adaptability front of mind, to ensure they are able to leverage the best AI has to offer now and are ready to implement novel technologies yet to come. 

“But adaptability should fit in with the governance fundamentals: organising your data, updating your SDLC to include data engineers and end users and investing in talented employees who are eager to evolve. 

“Vibe coding in the world of enterprise software will need to understand the notion of non-functional software requirements, governance and control. Until then, it’s a short-term con with limited long-term gains.”

Vibe coding promises to close the gap between having great ideas and being restricted by the lack of coding knowledge to turn them into reality. 

But, the current landscape needs to mature to adapt to fix bugs and develop vibe coding tools that solve real-world problems that can be safely integrated for it to be truly democratic.