Is vibe coding the future?

As developers have embraced AI to a greater degree, many have hailed the rise of so called vibe coding, in which broad prompts are given to AI models to produce refined code that's actionable and in many cases deployable.

But vibe coding also comes with risks, because by its very nature it's more of a hands-off activity. It decreases the degree to which developers can explain what their code is doing, and may lead to them overlooking vulnerabilities.

How serious is this risk, and what are the potential benefits of vibe coding in the future?

In this episode, Rory speaks to Simon Black, director of evangelism at Mendix, to unpack what vibe coding is and why it comes with such high risks.

Highlights

"I think currently, vibe coding and the process of vibe coding has some of limitations. I think the challenge of vibe coding is that they are trying to speed up the part of development that people find most difficult, which is writing the code. And you know that that comes with a lot of advantages. You can get things done very, very quickly, but it often forgets about all the other challenges with software development, things like being able to do peer reviews, being able to improve the deployment, the full software development life cycle."

"You need to need to give it context of your organization and how it builds software. Not just the code but how do I evaluate what type of software we should even be building as an organization? How do I assign the value to it? How do I understand which projects to pick first? You need to be able to give it context around how you govern, how you audit projects."

"They've taken a coders' approach to a no coders' world. Let's say it makes sense. So they've gone, okay, well, we need to build applications quicker. The best way we can do that is by giving AI the ability to generate code. Okay, great. That works if you're giving to a developer audience, not so great if you're giving it to maybe more of a citizen developer, or someone that is not from a software development background, and expecting them to understand, you know, TypeScript or understand React code or understand how databases work and so on.

Footnotes

• Senior developers are all in on vibe coding, but junior staff lack the experience to spot critical flaws
• Want to supercharge your vibe coding skills? Here are the best AI models developers can use to generate secure code
• AI coding really isn't living up to expectations – "the savings have been unremarkable" but not for the reason you might think
• Researchers tested over 100 leading AI models on coding tasks — nearly half produced glaring security flaws
• MCP servers used by developers and 'vibe coders' are riddled with vulnerabilities – here’s what you need to know
• The Replit vibe coding incident gives us a glimpse into why developers are still wary of AI coding assistants

Subscribe 

• Subscribe to The IT Pro Podcast on Apple Podcasts
• Subscribe to The IT Pro Podcast on Spotify
• Subscribe to the IT Pro newsletter
• Join us on LinkedIn