I think the argument is that if your neighbor's couch catches fire, there is a risk of the fire spreading to your apartment.

I've seen a number of fire videos as part of a training class. It is astonishing how fast a couch + room full of furnishings will burn: https://www.youtube.com/watch?v=ult6Biaf7oM. From ignition to flashover/fully-involved in 2-3 minutes. In a major urban area like SF, 5-10 minutes is a good response time for the first units on scene, not to mention the time to coordinate a fire attack, ventilate and flow water.

I don't know whether the claims in the article are well-supported or not (or what type of treatment the couch in the video had, as the videos are from the late 80s/early 90s), but I think most people don't realize just how dangerous a working fire in a single room full of ordinary furnishings can be.

That's where I think droithomme's comment comes into play. Where the downsides are so high, strong positive evidence should be provided and I'm not convinced it's there.