Postgres documentation says this explicitly:

If ORDER BY is not given, the rows are returned in whatever order the system finds fastest to produce.

"The system" in this case comprises the postgres daemon itself (including implementation of its data access methods and the query optimizer), the underlying operating system, logical and physical layout of the database storage, possibly even CPU caches. Since you as the database user have no control over that stack you should not rely on it continuing to behave forever the way it behaves this very minute.

Your colleagues are committing the hasty generalization fallacy. To disprove their point it is sufficient to show that their assumption is wrong only once, e.g. by this dbfiddle.

This is the black swan story all over again. If you haven't seen one yet it doesn't mean they don't exist. Hopefully in your case it won't lead to another world wide financial crisis, simply to a few unhappy customers.

Postgres documentation says this explicitly:

If ORDER BY is not given, the rows are returned in whatever order the system finds fastest to produce.

"The system" in this case comprises the postgres daemon itself (including implementation of its data access methods and the query optimizer), the underlying operating system, logical and physical layout of the database storage, possibly even CPU caches. Since you as the database user have no control over that stack you should not rely on it continuing to behave forever the way it behaves this very minute.

Your colleagues are committing the hasty generalization fallacy. To disprove their point it is sufficient to show that their assumption is wrong only once, e.g. by this dbfiddle.