The sentences of the form 'Group A is bad', when presented out of context, are ambiguous between
Group A, considered as a group (i.e. collectively) is bad.
Members of Group A, considered individually, are bad.
If what one intends is of the former kind, no generalisation is involved. One may, for example, say that a political party is bad on the ground that its announced programme aims at something bad. That does not imply anything about whether the members of the party, considered individually, are bad: it may be, for example, that many of them joined the party because they were confused, misguided, or manipulated by propaganda. Or one may say that an army fighting in some war is bad, on the ground that its military aims or ways of fighting are bad. Again, this does not imply anything about the badness of individual soldiers: they may have been drafted.
If what one intends is of the latter kind, another ambiguity will appear. Saying that the group's members are bad can mean that all of them are, or that many are, or that a significant number are, or that the typical members are, and so forth. A careful writer/speaker will strive to make it clear which of these is meant, and would not just say that the group is bad, without qualification. Now, if one says that all members of the group are bad, and one hasn't actually examined all of them, then, yes, a generalisation is involved: one is generalising from the members one has examined to the others. Similarly, if one says that many members are bad, but has examined only a few, a generalisation is involved.
If what one says is based on a generalisation, there will be a further question of whether the generalisation is justified. The generalisations that people make in casual conversations are often unjustified, but if the generalisation is based on the examination of a sufficiently large sample of the group's members, and the sample is representative of the group, the generalisation may be justified. What makes the sample sufficiently large and representative to justify a generalisation is a matter of logic (broadly conceived), scientific method, statistics, and epistemology; it is not a matter of the language, and is thus outside the scope of this site.