The biggest pitfall of vibe coding is that you don’t know what you don’t know. This allows AI to lead you down the wrong path.

I’ve noticed an increase in people contacting us to change settings in their Google Workspace after AI prompted them to do so while vibe coding an app. The problem is that almost every time this comes through, the AI has provided incorrect information or an inadequate solution to what they want to achieve.

Over the weekend, a client contacted us to request changes to some Google Workspace settings for an app they were developing. Changing the settings the AI requested would have created a major security vulnerability across their Google Workspace.

This is the problem: you don’t know what you don’t know. If you rely on the AI without giving it adequate prompts, it may produce something that works but could also contain fatal flaws. The flaw was that the client didn’t ask the AI, in their prompt, to create a secure solution that complies with privacy policies.

One simple way to avoid this is to include specific requirements in your prompts, such as asking for a “secure solution” or making sure the result is “privacy-compliant.” By adding these details to your requests, you’re more likely to get a response that considers important aspects such as security and compliance.

If you go back to the days of Googling, you had to know something about the topic you were researching so you could tell whether the page you landed on was accurate. The problem with vibe coding is that it isn’t clear, especially if you’re unfamiliar with the system you’re vibe coding for and have little coding knowledge.

This reflects a broader issue with AI: the AI will only act on the prompt you give it, and if the prompt lacks vital information, the AI won’t point out what’s missing - unless you specifically ask.

And that’s why partnering with a leading AI IT support partner like Kimbley IT is vital so we can identify these risks before you vibe code them into production. Our guidance helps prevent costly mistakes, protects sensitive data, and provides peace of mind knowing that your systems are built with best practices in mind.

So how did I help the client at the weekend? I didn’t deploy the changes they asked for; instead, I guided them down the correct path and produced the prompt they should have asked their AI for if they had known what they didn’t know. They are now on their way to creating a really handy app for their business. 👍

Here’s a photo of a friendly juvenile robin that came right up to me in the garden over the weekend.