Stop. Just stop saying that soon ____ will replace dogs. Nothing and I mean nothing will replace dogs.

Here’s the latest, a “robot bloodhound.”

I have no idea how much money is poured into efforts to replace dog noses with machines that detect:

• Cancer (multiple kinds)
• Low blood sugar
• Other medical conditions
• Narcotics
• Bombs
• Contraband, including veggies in a traveler’s suitcase
• Contraband, including smuggled animals, plants, or anything else in shipping containers
• Scent trails on land
• Scent trails on water
• Cadavers (on land or under water)
• Identify human cremains
• Endangered / rare wildlife
• Endangered / rare plants
• Insects
• Rats
• Other pests and parasites
• Truffles

I could literally list millions of things that dogs can be taught to identify and find by scent. Sometimes, humans don’t even know what the dog is scenting. But dogs are so smart and capable that we can teach them to reliably find it anyhow.

To replace the dogs, a machine would need to be able to identify the exact chemical combination that creates the scent. The machines, so far, are less accurate than dogs. This might be because identifying a scent is not a cut-and-dried, easily reproduced, often repeated set of identical steps.

It’s a process that requires thinking and intuition and understanding of the goal. Dogs can do all of that. They do it all the time, every, in their interactions with us — even without training. The machine does exactly what the human programmed it to do. It might be able to “get better” with practice, but only within parameters determined by the humans who programmed the algorithm. Regardless of what you might hear about how smart computers are, they are not thinking on their own.

Dogs are.

And, dogs can go into all kinds of situations, flexibly adjust to different working conditions, offer feedback on what’s happening, and learn from their successes and their failures. They can work in any kind of terrain and in weather that would defeat many mechanical imposters, uh, substitutes.

Additional dog teams can be trained far less expensively than fancy schmancy robots can be built and moved around to world to anyplace they might be needed. Dogs can hop onto a plane with their handlers and go help — at a disaster site, in a large-scale search-and-rescue operation, at a field hospital.

Remember, scenting is only one of dozens, maybe hundreds of ways dogs assist people. Or make our lives better without a specific task — just by being dogs.

So just stop it. Stop pouring millions of dollars into machines to “replace” dogs. Focus that money and effort on training dog-and-handler teams. That could prevent things like what I heard on the radio this morning: In a story on the return of remains from North Korea, which are likely to include the remains of many American servicemen who died there more than 60 years ago, a comment was made about how difficult it is to find human remains because there’s no technology that can detect them. Maybe there’s no technology, but trained cadaver dog teams could certainly find them. They wouldn’t falsely alert on bones from nonhuman mammals, either, preventing the “return” of nonhuman remains (which has happened).

And remember, each of those expensive robots does only one task. A different robot is needed for each task. The dog team? That dog can be trained to do multiple related (or unrelated) tasks. Besides, who wants to cuddle up with the robot bloodhound at the end of a tiring day of searching?