The chances of retinal damage would appear to be low?
Just to caveat this:
I am an engineer - not a doctor and not a magician, so all I will ever talk about is whether there is a risk that received power levels approach or exceed the values that SOMEONE ELSE has published papers saying they are dangerous.
So I cannot say "The chances of retinal damage when lasered above x,000 feet are low" - all I can say is that "The indications are that when lasered above x,000 feet the exposure won't exceed the values which the published papers say will cause retinal damage".
Now based on the sums I did last night the indications are that the risk of the retina receiving power levels above that threshold from a ground-based laser pointer while flying above (pick a number - let's say 5,000 feet) is probably very, very low, but the risk at low level (again, pick a number - I choose 500 feet) is probably high enough to be a concern. But it would still require very clear air and probably require illegal devices and very accurate targeting.
It got late last night and I didn't finish what I wanted to say. So to continue - why did I mention the 10sec/1sec/0.1sec exposure times? Well when I first raised the issue of the targeting accuracy needed[1] in the thread I'm now excluded from it was countered by the suggestion that the beam wasn't "aimed" but "waggled about" [with some gratuitous suggestions about the waggler being drug-crazed, and maybe even a Trump supporter]. So again I did some math to evaluate this which was deleted because I included opinions. So I'll try again:
If someone has a pointer and is just "waving it about" the angular rates are likely to be of the order of more than 60 degrees per second (try it). Based on data from targeting trials (not public domain - sorry) for someone trying to aim at a device without any sights the "shake rate" is likely to be in the region of 20 degrees per second, dropping to 10 degrees per second for plain sights (people can get much better than this with telescopic sights and lots of practice, but neither has been suggested for morons lasering aeroplanes).
So what does that mean? Well for the aeroplane at 8,000 feet and three miles range a 10deg/sec "waggle" translates into the beam moving from side to side at 2,800 feet per second. We said the beam would be 10m (lets call it 33 feet) diameter, so any single point (like an eyeball) will only be within the beam for 12 milli-seconds: 0.012sec or one eighth of the fastest "flash" case I discussed in the previous post. So even if we pick some very pessimistic assumptions (clear air, high power, good laser and an aimed laser with sights) we are still many orders of magnitude too small to get near these published damaging power values.
Looking at the 300 foot case it suggests that the seemingly "dangerous" values mentioned above need to be divided by eight for a real-world case - or in other words the danger threshold power values are closer to 1.7W, 3.4W and 6.25W (or 1700mW, 3400W and 6250mW if you prefer) for the 10 second, 1 second and 0.1 second exposures respectively. We're start to get up to power levels which are harder to find on the open market (or even ebay!) even for the low-level case. Although the flash-distraction would still be an issue of course, and I can't find any particularly digestible authoritative data on the power levels at which flash after-images become persistent for more than a few seconds so I can't include that in the anlysis.
PDR
[1] Where someone said "laser pointers are used to point at stars, so obviously it isn't a problem" - huh? If you pointed a laser pointer at the nearest star it would be 9.4 years before you saw the dot even assuming you were pointing it in the right place! Do people never think about what they are saying??