Regarding the 3 months: the UK only just changed it from a lifetime ban for MSM to 12 months - they feel that 12 is safer, many people have argued that this is still over the top. I'm not a doctor, and I'm not well enough informed to tell you who is right on this. Countries that do have a lifetime ban certainly need their policies updated.
Regarding unprotected: the argument is that even protected sex isn't 100% safe. I don't have any stats to argue for either side here, and I don't have an opinion for that reason.
So, it's not entirely a case of "we can't be bothered to test", but at the same time many countries have out of date policies. And even those, like the UK, with better ones, there are still some people who argue they are too strict against MSM, and others who argue that any less strict would be too dangerous.
I could very well have had sex with 100 women in the past 3 months (as a man) and been unsafe with all of them. Yet, they don't seem to be the least bit worried about that scenario.
Having 100 random sexual encounters with randomly sampled women leaves you approximately a 50% chance of having unprotected sex with someone who has HIV. Not to mention, HIV may not even be transmitted even if you have an encounter.
On the other hand, after 100 random homosexual MSM encounters, your chances of having had unprotected sex with someone carrying HIV is 99.99997344%.
MSM puts you at a much higher risk for HIV. That's fact.
This seems a lot more sane than "There's this group that we used to consider high risk for HIV. We're not willing to scientifically test for it. Are you a part of this group?"