True, but the alternative is tungsten, which is not much better in terms of toxicity:

https://www.atsdr.cdc.gov/toxprofiles/tp186.pdf

The problem here is that armor piercing rounds need to be made of heavy metals in order to work and heavy metals are generally toxic. Even lead, which old fashioned bullets and shells are made of is toxic.

Uranium is more likely to distribute that toxicity in a convenient dust cloud, though. Tungsten and lead for the most part stay in one piece on impact.

This is in the context of a substance we’re trying to use to destroy enemy tanks and kill/injure combatants, not make s’mores out of.

Though it remains in the environment even after the operation and civilians will be exposed to it. This has been the case in e.g. Balkans (https://www.nytimes.com/2001/01/07/world/radiation-from-balk...) and also in Iraq (https://www.theguardian.com/world/2014/jun/19/us-depleted-ur...)

Perhaps I should have said "not outrageously toxic" or something of that sort, because that does not sound worse than the stuff we make normal bullets out of - lead.

Lead doesn't shatter the way uranium does, so it's less inclined to form a breathable dust on impact.

That's true! Lead does tend to stay more clumped up than uranium dust. But isn't it also easier for lead to become bioavailable and/or enter the water supply?