s smelling antenna in your location and then some criminals come in and they do some crimes but there's no video. Would it be possible to use the electronic nose to then get a sample from the person should you ever catch them and see if they're guilty based on their smell? Because you may have recorded their smell at the location and there would be... it's probably like a fingerprint. So there's a whole bunch. If you assume that smell is, for humans, the underappreciated sense — you know my dog has super smell — if we had robots with super smell, I feel like a whole bunch of things would happen that you don't see coming. So I think it's a big deal.

And you should trust science. I encourage you to always believe science. Whatever science says, it's all true if science says it.

Our next story is from *Nature*, the publication. Turns out that the journals with the high rates of suspicious papers... there's this startup that has some kind of mechanism for determining which of the scientific papers are bogus. It's called Argus. That's the name of the startup or the system that looks for research papers that look like they're not good.

Let's see how many do they find? I mean, there can't be that many, right? Because we all trust science. So I mean, sure, nothing's perfect, but if it's science, probably... I'd be worried they might find like a dozen papers or so that are just made up. I mean, imagine how shocking that would be if they found like maybe 20 science papers that were just made up. That would really rock your confidence, wouldn't it? But imagine if it was like 50. What if they found 50 science papers that had been accepted as peer-reviewed and then turned out to be just totally false? I mean, you can't... see how many was it? They flagged more than 40,000 high risk and 180,000 medium risk papers, and they've indexed more than 50,000 retracted papers.

Okay, that's a little worse than I thought it was. Now these are just flagged, the 40,000 high-risk, but the tool is saying that there are 40,000 papers that you probably shouldn't count on.

Now let me ask you this. Suppose I created this system on paper and it never existed before. It goes like this: If you want to be a scientist, you've got to publish papers. Well, all right, so far so good. Yeah, that makes sense because then you can tell if the scientists are really up on their field and stuff. Good. Yeah, scientists should be able to publish things. We're going to have a peer review. There'll be other scientists who will look at it, not to totally make sure that's fine, but to weed out obvious quacks and stuff. And I think to myself, yeah, that makes sense. That does make sense.

Yeah. And now we're going to give an incentive structure to the scientists that they will make lots of money if they publish, but they won't make lots of money if they don't. Oh, okay. Well, yeah, I suppose anything that you want to be done, you want to have a little financial incentive for sure. So sure, sure.

Now, how do you think that's going to turn out? Well, if you dropped a weasel into that situation, they would say, "What if I just make up a bunch of papers, make them look like they're good, send them to my buddy who's the cousin of my brother-in-law to do the peer review? He probably won't even look at it. Or I'll send it to somebody who always says yes, and they'll definitely get peer-reviewed, and then I'll get huge bonuses and I'll be paid to speak at events." What would stop you from doing that? Well, apparently not much, because you don't hear about people who are shamed forever because their papers were retracted. So it seems to me that even on paper, what that should have led to, given the financial incentive, eventually all science will be fake. Because the fake science is so e