Yes, except there have been arthropods larger than humans back in the day, the 8 foot long eurypterids, and there are still very large crabs out there in the 30 pound range.
Back when oxygen levels 300 million years ago there used to be lots of very large land arthropods. So the problem for arthropods isn’t gravity, it’s oxygen. They just don’t have the efficient lungs of vertebrates.
But if we’re imagining species that evolved on other planets that only superficially resemble arthropods there’s no saying what sort of lungs they’d have, or what sort of atmosphere they’d evolve in.
Anyway, the idea that if a species looks looked sort of like a scorpion they’d be cruel and vicious, whereas if they looked like a koala bear they’d be friendly and sweet is just silly. Dolphins and sharks look a lot alike, but are very different behaviorally. So expecting a creature to behave like a shark because it superficially resembles a shark is very silly.
Well, no, they wouldn’t, and no, it isn’t. There’s no reason an alien would have the same proportions as humans, so they could conceivably be much smaller than humans, but with a proportionately greater amount of their body mass given over to mental functions. And there’s also no reason to assume that a brain needs to be human-sized to support intelligence. That might be a size requirement for mammalian brains, but we’re not talking about a mammal, or even a true arthropod. We’re talking about something that merely resembles these sorts of creatures. They very well might have evolved a much more efficient brain structure, that allows for advanced intelligence in a much smaller package.
In one of his novels, Vernor Vinge described a race of genocidal aliens as “butterflies in jackboots.”
You could argue that humans ARE the intelligent reptiles of earth if you follow ancestry back 300+ million years to common tetrapod/amniote ancestors. Ancestors of crocodiles split off in one direction and ancestors of humans split off in another direction… but an animal like dimetrodon is on the human side of that split, not the crocodile side.
Even if you want to insist on a species derived from a modern reptile: if we suppose that an iguana evolved enough to develop technology for space travel, then it hardly seems necessary to assume that they must remain scaled, endothermic, egg-layers that look like Bossk.
If you’re thinking of a hive-mind, each individual part is not intelligent but all together they are. The individual parts could be any size.
Humans are a combination of a lot of living cells that work together. We just tend to be stuck together in one lump. It’s certainly possible a being could evolve that was a bunch of living creatures that worked together but weren’t physically stuck together. It’s possible to think of a bee hive or an ant nest in that way.
Crows are very clever, playful and relatively small. As direct descendants of the dinosaurs they are much closer to the concept of an intelligent dinosaur than the bipedal reptilian fantasy of ‘crappy humanoid dinosaurs’ mentioned in John Mace’s link upthread.
If something with a brain as small as a corvid’s can be smart, then there really doesn’t seem to be much evidence that intelligent aliens need to be roughly human size, or need to have a shape anything like a human. Olaf Stapledon explored a wide range of possibilities for intelligent lifeform types in his book Starmaker; one intriguing possibility was a flock of birds, each with fairly respectable individual intelligence, bur capable of behaving as a larger superorganism when necessary.