Hopefully we don't. But we need to solve dementia which might require a deeper understanding of the connectome. And furthermore, the longer we age the more new diseases we will discover.

By the time we have multi-centenarians there might be whole new types of dementia that could arise.

Perhaps a better question than (do we need AGI) might be, do we need to solve protein folding before we can make meaningful progress. ie should protein folding be considered an ancillary problem or a perquisite for other solutions that must developed on the technology stack to make progress in age research