Are laser diffuse reflections really so dangerous? Class IV lasers are lasers with an optical power higher than 500mW, and for these lasers "even diffuse reflections can cause permanent eye damage". Well, I would like to understand how this is possible exactly.
Let's imagine I have a 1W laser pointer and I shine it on a white wall. If the wall has a rough surface, then the subsequent scattering of the light should appear approximately isotropic. Let's say it is, for simplicity, and let's imagine that no light is absorbed by the wall.
In this case, then, 1W of light will be emitted from the wall in a semisphere around it.
Let's now say that I am at 1m distance from the wall and I look at the reflection. Let's say that my iris is 10mm in diameter (this is more than normal for a human), which means that I should intercept about [π(0.01)²]/(2π)=50 microwatt of optical power... How can this be dangerous for the eye?
Am I using the wrong assumptions or calculations?
 A: 
imagine...I shine it on a white wall.

How small is the spot? With a lens to focus it, you can focus all of that energy onto an incredibly tiny spot.

Let's now say that I am at 1m distance from the wall ... I should intercept ... 50 microwatt

OK, but the people who write the safety standards are more interested in protecting you from the worst thing that could happen, not the thing you were hoping would happen. So, let's say instead, your eye was maybe 10cm from the spot while you were adjusting something on your optical bench. Now we're talking about 5 milliwatts.
If your eyesight is sharp, the entire five milliwatts that enters your eye from that incredibly tiny spot could be focused onto a single photoreceptor in your retina.
I'm not going to enter in to a discussion of what that five mW could do when it is delivered to a single cell, but that's closer to the scenario that they're trying to warn you about.
A: Your calculation is fine, but it makes some assumptions that would be an unwise foundation on which to rest your eye safety.

*

*You assume that you are 1 meter away. This might be true most of the time, but you only need one mishap to permanently damage yourself. You might not always know where the beam is, or you might forget and lean in to get a closer look at something.


*You assume that diffuse reflection is isotropic. It is not, in general. There was a time in grad school when I was playing fast and loose with laser safety (in retrospect, we had a poor safety culture), and I saw a bright flash of light while I was standing at the optical table above a 1.5 W, 780 nm laser. It wasn’t enough to damage my eye, but I could tell it was close. I thought I knew where the beam was, but there was an unexpected diffuse reflection that scattered a large fraction of the light up from the table with (I’m guessing) a ~5 degree divergence angle. It was a close call I have not forgotten.
