And there might be a person specific id in the link you get, which other people copying the image don't get.
For example edge did for a while scrap any link you visited with it, even if you logged out of exactly this feature, not running any heuristics about weather it might a a magic link and then skipping it and AFIK also not stripping query parameters (through it did respect robots.txt I think).
Similar a lot of apps do similar, sometimes not even respecting robots.txt.
Through in any case it's still Meta being rediculous, because it's a known issue and you can't fault your user for doing perfectly normal things in a world where even one of the major browsers behaves like malicious spyware wrt. urls/web scrapping without user explicit consent.