Yes, some vehicle OEMs spec synthetics, but we have to ask why they do so.
Excellent example (one I learned of from this site a few years ago) is where Chevrolet uses M1 in the Corvette from the factory. In high-perf applications, such as racing and weekend ultra-hard runnings like auto-cross, the 'Vette would signficantly heat the oil. The M1 was able to withstand that abuse a bit better than conventional oils. The lube cooling system was only able to draw down the heat so much, and after that, it was important that the oil be able to tolerate a higher heat load. Hence, to protect the engine in the most exreme circumstances, GM spec's M1 from the factory. Of course, that means NOTHING to the 76 year-old man that drives his 'Vette to the club for a round of golf. He NEVER gets the oil hot enough to warrant a synthetic, but it's spec'd, so that's what he uses. Look around you; there are a LOT of older men and women driving 'Vettes (presumably reliving their youth). Do they NEED synthetics for their short OCIs and puttering around town? NO! For people that use their 'Vette in hard racing apps, then I'd agree that synthetic is a great idea. But that means nothing to the folks that only scoot around town; they don't "need" synthetic, even though GM spec'd it. In fact, some could argue that because GM couldn't appropriately design the lube cooling system, the synthetic was a necessity to overcome an engineering deficiency. Two viewpoints to the same issue. I'll not take sides; just pointing out how there is more than one way to view this topic.
Then consider the opposite effect; I recently learned of the propensity of Saturn SLs to coke the oil rings due to a design flaw. GM spec'd dino oil, and perhaps should have spec'd M1 in this case. But they didn't. And in "normal" applications, even dino is a risk at OEM OCIs. Same goes for the infamous Toyota sludger engines from many years back. Of course, if you shorten up the OCI to some unknown factor, you're likely to avoid the sluge issue. Overall, though, both GM and Toyota ultimately failed in their prediction. Here's the point:
in LAB testing, I susepct these conditions were proven to be true, but in the real world, the oil/OCI combinations failed. AZsynthetic would have us believe that every single OEM spec based upon lab testing is flawless; clearly that isn't the case. I certainly do believe that most OEM specs are valid, but it is also true to say the not all are so; there are failures in the field that cannot be well predicted in the lab.
Further, IIRC, Mercedes Benz actually states that, for some applications, regardless of what oil is used, the OLM is to be followed. It was brought up about a Sprinter, or some other vehicle with a MB engine. The manual specifically stated that regardless of conventional or synthetic oil use, the OLM was the limit; no more. The OEM acknowledged that synthetics would not be offered any more OCI than the dino. This is a direct contradiction to what AZsynthetic would have us believe; that all OEMs know synthetics are superior. I have not personally seen any broad-market based owner's guide (Ford, GM, Dodge, Honda, Toyota, etc) that specifically stated " ... if you use synthetics, you can double your OCI ..." or something to that effect. If it's out there, I have yet to see it. That does not mean that it does not exist; I admit I don't have access to all the data in the vehicular world. But the MASS MARKET EQUATION shows that OEMs simply ignore the dino/syn topic in the owner's guides. If they have such knowledge that syn's are "better" as AZsynthetic would claim, then why are the statements not out there for the masses to read? I don't see such evidence in the owner's guides, the maintenance manuals, the OEM websites, etc. Please, point out the mass market evidence that OEMs support the superiority of synthetics. For example, my Fusion specs Ford WSS-M2C930A oil; the Motorcraft lube meeting that is a semi-syn. And there are many syn's that meet that. But there are also a lot of dino oils that meet that spec too. And my UOAs are proof that any lube meeting that spec is going to do more than well enough for the OEM OCI. Ford does not care if it's dino, semi or full syn; it must meet their spec for the warranty to cover OEM OCI. And they state nothing about OCI extension or "better" wear if you use a syn.
My point:
OEM make recommendations based upon expected conditions they want to warrant, and given their engineering knowledge at the time. No more; no less.
UOAs, OTOH, give direct knowledge to the individual based upon the descrete and distinct personal operation of the equipment. UOAs can also be used by other persons to judge their overall success or failure as compared/contrasted to the programs of others. It is not a complete answer, but it is by far the cheapest alternative we have for predictive behavior.
I'll take a UOA from real world use over the OEM lab-derived spec any day. The OEM spec is a great place to start, but it is NOT the end-all, be-all answer.
The two most grossly misunderstood topics when it comes to statistics is micro versus macro modeling, and correlation versus causation. This is what I do for a living. I'm confident we can use UOAs for comparitive purposes, as long as we understand the limitations of our testing protocol.