I think the problem isn’t when you ID something and don’t know what it is. It’s when more than several people agree with an ID even though they aren’t sure that it’s the best answer.
I know some of the privets very well. Specifically, I know four species found from Texas to the Atlantic Coast. Often the first person to identify it gets it wrong—either it isn’t a privet at all or it isn’t the privet they said it was.
If they’re the only one who made the wrong ID, it’s easy to fix. I add the correct ID and usually tell them the most obvious difference between the best ID and the ID they made. They quickly fix the problem, learn a bit, and continue observing, identifying, or both.
Even if they tune out—they made the ID, but they’ve decided that iNaturalist isn’t interesting and never return—it isn’t a big problem. If I can bring in two more people who either know privets well or know they can trust my judgment, then we can make the observation become research grade with the correct ID.
The problem arises when too many people who aren’t sure that the initial ID is correct agree with it. Everybody else thinks it’s Japanese privet, and I know it’s glossy privet, but the page says it’s a research grade ID of Japanese privet, and I’m labeled a maverick. If others who come to the page agree with the first ID, it gets only worse. To swing the ID to the correct one, we have to come up with twice as many people who got it wrong, plus one.
So it’s OK to be the first to chime in when you aren’t sure—but try to be a little more careful about agreeing with an ID. If you are sure that it’s right, go ahead and agree. If you notice something about the observation that confirms the ID but hasn’t yet been mentioned, point it out. That’s how you help the community learn.
On the other hand, if the ID you see just happens to be the same guess that you would make, then don’t agree. Instead, follow that observation. Ask any questions you have. That way, even if the first ID turns out to have been right, you’ll learn more about the species in question, and perhaps even about its lookalikes.