Many of the people I met who were considered experts and had a vast depth of knowledge never referred to themselves as experts and seemed do get embarrassed at the idea.
Whereas, i've met several self-proclaimed experts who liked to talk a lot about how great they were, but never really showed the depths of knowledge the other kind of experts showed.
At least from my experiences anyway in life.
I think the correct criterion is neither of these: it's whether the expert can back up their expertise with models that have a good predictive track record. That's the real test of whether they actually know what they are talking about vs. just blowing smoke. Unfortunately, this criterion is hardly ever applied that I can see.
The problem is, in those areas where there's not a tight feedback loop as mentioned in the article, it's possible to build your way to the top by just being somewhat skilful in the field and knowing how to appear confident, and how to (over)sell your work, and knowing when something is technically right but you should not say it because politics. And on your way to the top you're going to out-compete those that are very skilful on the field, but don't have such good communication abilities.
I dislike interviews, but I like having a job.
All of that goes out the door with pundits and shills, who don't just appear once, but have chosen the field for fame/fortune... and for those who have largely chosen to sell their research topics/results for profit. However, those are relatively easy to spot by most viewers and journalists who care.
Indirect benefits are harder to evaluate and will be nature involve a bunch of hand waving, but there is a huge difference between say holding a research chair, and having a consulting business as an "available expert".
I don't think there is a meaningful difference. First, I don't believe people can make choices in any kind of isolation. If you believe you are making a choice without your ego involved, you have simply deceived yourself. Congrats, you have a high functioning ego. Second, even if they could, audiences are not capable of concluding with any reasonable certainty which is motivating someone appearing on television. I don't mean that derisively, like people are stupid, I mean it is essentially beyond humans to do this.
IMO, in the absence of actionable information, it is pointless to speculate. Basically, you cannot know it so it's not worth thinking about.
I guess my hypothesis is that you get better information from people whose reason for agreeing to interviews is not self promotion. If true, that's actionable.
I suspect this misses some nuance; there is a meaningful difference between people who appear on the media because they want exposure, and people who do it because they want their field of study to be well represented.
For example there are people to accept interviews when they genuinely dislike the process, because they felt it was a duty that came along with their chair, etc. I doubt these are ever the most dynamic interviews, but they are common.
I agree this doesn't really extend to the level of the same old bobble heads pulled up on prime time news TV, but don't thing you can paint all of the media exposure with this brush.