Then their warranty should cover damage caused by splashes or being caught in the rain (potentially very different things, by the way).
Nobody is saying that because they show someone being caught in the rain, the warranty should cover sub-aqua instagramming...
Meanwhile, they did use a picture of someone falling into a pool holding a phone - but apparently that's OK because it wasn't an iPhone X and was taken before the phone actually entered the water so using it doesn't actually imply that the iPhone X could survive a fall into a pool and a strange invisible man I met in a bar did it and ran away...
If they just specify that its IP68 then they should warrant that it meets IP68 - no more, no less.
However, there seems to be a side-problem with IP68 being easy to misinterpret (possibly by design) and representing a set of circumstances that don't really reflect real life (...e.g. being dropped into salty/chlorinated/detergent-filled water by someone who hasn't memorised the manufacturer-approved post-immersion drill) so maybe they shouldn't give it too much weight as a feature.
This thread is about water-resistance claims. Other threads discussing BendGate, CrackGate, FlexGate, HoldingItWrongGate etc. and whether or not Apple should fix them under warranty are available.
I think a "reasonable person" understands that toughened glass does not mean "unbreakable" and I don't recall offhand seeing Apple (or Samsung) claim any more than that. I don't plan to waste my morning hunting for claims that Apple and/or Samsung may have made about break resistance - but whatever they claim, they should warrant (...and, in many jurisdictions, could be held to what they claim - plus a general requirement along the lines of being 'reasonably durable' for that class of product - whatever the warranty says).
This really comes down to what you or I believe a "reasonable person" understands. You think a reasonable person understands that glass breaks, but that same reasonable person sees water resistance differently. I think a reasonable person understands the two are little different.
Maybe it's a matter of age? I grew up in the age of mechanical watches. It was settled law at that time (in the U.S.) that a "water resistant" watch is different than a "waterproof" watch. "Resistant" meant (and still means) it is built to resist water intrusion, but that the risk is the owner's. The owner's expectation is that a bit of water now and then probably won't hurt (washing hands, accidentally wearing in the shower, etc.), but if they go swimming or snorkeling, all bets are off. "Waterproof" is a manufacturer claim that water can't get in (well, so long as the crystal, case, and crown remain intact), and generally is covered under warranty.
Maybe I'm not a typical "reasonable" person, because my personal experience extends to whitewater canoe/kayak and hiking/backpacking. Whether it's full immersion in a river or an occasional day of hiking in the rain (or nights in tents during downpours)... water usually wins, even when using so-called waterproof gear. Brand-new gear tends to perform as advertised, but after some use things begin to break down. Some cleaning/maintenance, maybe lubrication of o-ring seals and patching of punctures will keep things going for a while longer, but that's about it.
That's the same expectation I bring to electronics. Again, my background may make me atypical, as I've done a fair amount of electronics repair in my day. I know how the stuff is built, and I certainly know what even a tiny bit of water can do to an electrical circuit. Truly "waterproof" electronics (the kind found on ships) is built to a very different standard.
So I see the water resistance of an electronic device to be little different to these other examples - that it's great that it's built to withstand some exposure to liquids, but that water can still be expected to win. Defenses tend to break down over time and must be maintained, while water's properties never change.