I'm amazed my position is even remotely controversial here. Many people here, including yourself, are framing the debate from the perspective of their own experiences as candidates, which is a biased way to look at the perspective of an interviewer.
The fact is interviewing it tough, and is inherently a flawed process (I agree with you that ideally the interviewer could get to know the candidate's character, but an hour or a day isn't the correct timeframe to do this). Good interviewers try to find the best fit for their teams, instead of lamenting the fact that the problem can't tractably be solved perfectly.
No, actually I'm not. I've never had to embellish my CV and every interview I've done has resulted in a job offer.
I'm actually attacking the notion of "character" itself. Yes, there is a set of habits and probabilities of certain responses to various scenarios, but I contend that the mental model of "character" (as it is normally understood in Western culture) is a poor predictor because it groups together certain situational responses with labels that have no necessary underlying correlation to validate them.
My own judgement when interviewing job candidates mostly comes down to technical depth. I'm primarily interested in competence and demonstrated ability to deal with and get results in a complex technological system. I'm not really interested at all in CV buzzwords - and that's what most of those bullet lists end up being, tags to try and get through HR filters. But my experience is also coloured by working in relatively large companies, and being a technical guy, not a manager making the hire decision.
The fact is interviewing it tough, and is inherently a flawed process (I agree with you that ideally the interviewer could get to know the candidate's character, but an hour or a day isn't the correct timeframe to do this). Good interviewers try to find the best fit for their teams, instead of lamenting the fact that the problem can't tractably be solved perfectly.