In general in this type of use, "clever" and "dumb" would better be called "tricky" and "obvious" respectively. Usually people describe very tricky solutions as "clever", and much more obvious solutions as "dumb" jokingly.

For example, storing some flag in the high bits of a pointer field of a struct is a "clever" solution, whereas having a separate bool field is a "dumb" solution. In most cases, the "dumb" solution is much more robust over time (less likely to cause bugs as the code changes and is modified by various people). Of course, the "clever" solution is necessary in some situations (very constrained environment, critical infrastructure such as an object header used for every type etc), but should often be avoided if possible.

What's important is that the way this is often presented is that more experienced people will prefer "dumber" solutions, as experience often shows that long-time maintainability trumps many small losses of efficiency. So using "clever" and "dumb" in this way is not at all intended to put down the engineer writing the more robust version.

When it comes to everyday software solutions you are usually aiming for obvious and clear, and often 'clever' is obtuse and opaque but definitely not always.

I think there might just be some vernacular nuance here though, maybe we can call it smart and robust, versus clever and opaque. Some problems are just difficult though, and if you get to work on that kind of problem regularly then that is pretty lucky.