The lower your max. interval, the more frequently you will have to see your card. This, in turn, increases your daily workload, in minutes of studying. So at first glance, it may seem like keeping it as high as possible is good.
An "interval" in Anki can be interpreted as "the amount of time between two subsequent reviews". But I want to reframe it: an interval is the maximum amount of time that your material can spend being forgotten.
Let's assume that you reviewed a card, and the next interval is 20 years. Let's also assume that you immediately forgot the material in this card. This means that you will spend 20 years not knowing this material. That doesn't sound very good, does it?
At the same time, lowering the max. interval too much can increase your workload significantly. Here are the results of (somewhat old) simulations:
Average total time per card is basically "How many minutes you will spend on this card over it’s entire lifetime". "Anki" just means "the default algorithm of Anki".
As you can see, the amount of time spent increases as you decrease max. interval. Decreasing it from 36500 days to 365 days increases the total amount of time spent reviewing an average card by 50-60%, assuming you will be using Anki for 100 years.
So, should you lower your max. interval?
Well...maybe. I really want to find a mathematically optimal value of the max. interval, but I don't know how. On the one hand, I definitely don't want to spend 20 years not knowing the information in a card. On the other hand, I also don't want to do more work. Perhaps one day I'll come up with a way to precisely determine the optimal value of max. interval.
Until then, I set it to 5 years for most of my decks, except for decks with very hard material, where I keep the max. interval at 365 days. I think it's a reasonable compromise.