He was talking about laser pointers. Are they really dangerous?
Obviously the answer is that some of them are indeed a hazard, but I have to point out that his reasoning to determine which are safe is far from my own understanding of the topic, and advice from laser safety experts.
Dunning's approach was correct in pointing out that the human eye is more sensitive to green light than red, but his conclusion was the opposite of normal thinking. The reason might be that he has incorrectly assumed that increased sensitivity (i.e. ability to perceive the smallest amounts of light) correlates with sensitivity to damage (which it does not). The brightness of some green pointers is truly scary, but not all of them are actually dangerous.
Damage to the eye is caused by power, not by apparent brightness. OK - you might have a coloured spot in your vision for longer if the light seems brighter, but that doesn't necessarily mean that you have damaged your retina. On the other hand, infra-red lasers can easily blind you without seeming bright. Indeed they can do their terrible damage invisibly. (I work with people who operate IR lasers which would only be eye-safe at a distance of 20,000km by they way.)
In fact, the science in the above claims should be interpreted in this way. Green pointers appear brighter than red pointers for the same power. So a red 1 mW laser pointer seems surprisingly dim compared with a green pointer of the same power. To compensate for this, people are tempted to buy higher power red pointers. You might need tens of milliwatts to compete with 1 mW of green. This is where the danger lies.
Brian Dunning's conclusion that you should never use a green pointer is misleading and potentially dangerous. You should use green wherever possible, but just ensure that its 'peak output power' is less than 1 mW, not the 1 to 5 mW that Dunning recommended.
Then you can be certain that your audience's eyes will be safe. You should still avoid shining the light direct into anyone's eyes though.
Small note: There is a theoretical potential failure mode for green lasers, where they go to even higher power. If you spot your pointer getting much brighter then you should get it tested. In my own experience they seem to get dimmer with time, but this is only a matter of statistics (and perhaps ageing eyes?).