AI construction is becoming more time-efficient, but testing has to be given deliberate thoughts

Using AI-assisted coding tools, it has become possible to produce an element in minutes that used to take hours in the past.

While being amazing in its efficiency, such approach tends to give you a false impression that the work is done and it's high time to proceed to the next step.

As far as I am concerned, I have come to think of testing as being integral part of the development process rather than something that should come afterwards.

One cannot assume that the feature is done simply because the demo is working fine.

What is your minimum testing protocol when building with AI?