“This claim is true because the claimant speaks from authority.” Simple enough when you think of an example like “Objects of different masses fall at different rates because Aristotle said so.” Not so simple when the issue arises that it’s impossible to check every single claim yourself, and thus need to rely on some legitimate authority. That legitimate authority resides in making a logically sound case and building on facts obtained via scientific methodology with many different potential hazards analyzed, debated, and settled upon. In a word (or two): scientific consensus grants legitimate authority.
Why a scientific consensus rather than just whoever happens to be the smartest one in the room? Because of a foundational, ruthless rule of skepticism: Anyone is capable of saying anything. Nevermind that he has a degree or appear on TV or in a magazine. Even the greatest and most influential scientists can go off the deep end (read: Linus Pauling). But when ideas are argued against and have every nook and cranny analyzed critically by and skeptically by multiple people (training helps but is not sufficient) and still pass the test, that idea has legitimate scientific authority (since the process that gave it birth was the scientific method now, wasn’t it?).
Still, though, even once an idea has been put through the ringer of the scientific method and come out the other end, it could still be wrong. In this way claims about Nature are always liable for falsification further down the road, so scientific authority is never absolute. We can only say that we have such and such a confidence in this result based on the evidence accumulated thus far. That’s the world as revealed by scientific skepticism: a world of probabilities, of likelihoods, never of absolutes. “Have aliens visited the Earth?” “There’s no compelling evidence thus far, but I can’t be sure.” “Is general relativity correct?” “It’s passed all the tests so far, but we can never be certain.” You can never prove a negative.
And yet no single person has the time or energy to validate all the processes and methodologies of research going on (or that has already been done). That’s why we have the trappings of degrees and academic positions, to allow for some quick and dirty evaluation of merit. “He has a PhD in Astrophysics, so maybe his word on the likelihood of meteorite impacts has more validity than that dude on TV.” That can never be the be-all end-all of a truth’s claim though. A quick and dirty way of living in a quick and dirty world.
This fallacy can be abused when any result obtained by science is ignored, denied, or brushed aside since it’s “just an argument from authority.” Thinking about that for a second, you could use that to deny pretty much anything you want to and live in a totally sealed-off intellectual world. Individual people (scientist or otherwise) can say pretty much anything you can think of, and so you can never rely on somebody’s say-so when evaluating truth. In this way you can never be certain of anything, though practically your confidence level can be so high as to make the difference philosophical.