They didn't claim that there were any, just that AGI isn’t a necessary requirement for an application to be world-changing.
> There are possibly applications of existing AI/ML/SL technology which could be more impactful than general intelligence
It's not unreasonable to ask for an example.