As davy.j points out above, any oil used in a shared sump motorcycle engine tends to thin out/loose viscosity) over time. The EPA has pressured engine manufacturers to extend oil change intervals to reduce the disposal problem of used motor oil. But therein lies a problem. Tests have shown that oil that also lubricates the gearbox in a motorcycle looses viscosity quite quickly. The gears in the transmission are the significant factor in cutting the longer oil molecules into shorter pieces that are less viscous.
In one series of tests, non-synthetic motorcycle-specific oils had lost over 30% of their viscosity at 800 miles, and over 35% at 1,500 miles.
It should be noted that the viscosity of synthetic-based oils generally drops more slowly than that of petroleum-based oils in the same engine. Also, starting out with a 50 weight oil means that it takes longer to end up with a 20 weight oil, for example.
This doesn't speak well for 7,500 mile oil change intervals (also encouraged by the EPA).
So pretty obviously vehicles are designed to allow their oil to be within spec at some value substantially less than 100% of their 'as new' performance numbers, the million mile question is "HOW FAR under 100%". A quick search resulted in no mileage based testing of motorcycle oils for any model bike or brand oil that went beyond 1500 miles... which doesn't give us enough information to draw even a rough conclusion (pretty much useless actually). There seems to be a historical context driven 'wisdom' that 3000'ish miles is where the wear degraded performance of the oil falls below the 'safe limit' for <insert your vehicle of choice here>... but using a historical context where other factors confound the data doesn't allow for a trustworthy conclusion... basically it's your opinion at that point.
When someone says "Well I change my oil every 3000 miles, it's cheap insurance", I think fundamentally what they're saying is that they don't trust vehicle manufacturers to give them a data driven 'limit' for oil durability (including a reasonable safety margin). That's fine with me if it's your bag... but I've never seen any objective data that begins to indicate that any of them are misleading us. Ducati have defined a 9000 mile service life for engine oil on my bike, in the absence of data establishing empirically that they're wrong I'm going to have to take them at their word... because they're the experts not me.
Mobile 1 warranties one of their full syn's against oil based engine failure even with a 15,000 mile service interval (automotive). There is a LOT of data establishing that the 1500 and 3000 mile service intervals of the past were established (at least partly) due to leaded gas, and manufacturers increased recommended service intervals to the ~5k-10k mi range the same time they switched to unleaded.
IMO it's the same human idiosyncrasy that has driven tens of thousands of (otherwise) intelligent parents to conclude that getting their children immunized is a bad idea even though the VAST majority of competent Medical experts say it should be done - we get just enough information to form an opinion then stubbornly stick with it even if it has no valid data underpinning it. This is even more prevalent when there's a perceived risk associated with following expert recommendations. Luckily with engine oil (as opposed to immunizations) the only thing at risk is your wallet... so change away if it gives you peace of mind!