Ask an AI tool for twenty ideas - names for a service, ways to approach a customer, things to do about a quiet month - and twenty will arrive immediately, reasonably phrased and plausible. This is genuinely useful, and it is also the use most likely to waste time, depending entirely on what the person does next. The failure mode is recognisable: a list arrives, several items look interesting, the whole thing is saved to be considered properly later, and it is never opened again. Nothing was decided and forty minutes went somewhere.
The reason is that the shortage was never ideas. Most small business owners have more ideas than they can act on already; what they lack is the time to do any of them and a basis for choosing. A list of twenty plausible options does not help with either, and can make things worse by adding sixteen new possibilities to an existing pile of unfinished intentions. The tool is very good at the part that was not the bottleneck, which is a pattern worth recognising generally, because it is the most common way these tools waste time while feeling productive.
Used deliberately, though, generation has a real function, and it is a narrow one. Its value is in getting past the first three ideas, which are always the obvious ones and are usually the ones already tried. Somewhere in a list of twenty there are typically two that would not have occurred to anybody in the business, because they come from outside its habits. Finding those two is the entire purpose, and it requires going in having decided in advance what you are looking for and how many things you will take away - one, normally. A list you did not intend to reduce is a list you will not reduce.
The discipline that makes it work is to do the reduction in the same sitting, and to apply a filter the tool cannot: which of these could this business actually do, this month, with the people and money it has. That filter removes most of the list immediately, and it is knowledge nobody outside the business possesses. Then pick one, decide the first concrete step, and delete the rest rather than saving them. Deleting feels wasteful and is not - an idea worth doing will come back, and an archive of unexecuted possibilities is one of the more reliable ways to feel busy while standing still.