Do lasers always need something in the air to make their beam appear visible?

I think it’s really doubtful.
And, the laser isn’t the real problem - it’s the gigawatt hand-held power supply.

Ah, they’re 1920-gigawatt “death rays”.

Ordinary shop safety glasses generally don’t (although polycarbonate does block UV, so ordinary shop safety glasses might protect your eyes, to at least some degree, from a UV laser). As mentioned upthread, special goggles are required that filter the specific wavelengths of the laser you are working with.

Note that blocking ALL of the light from a laser may be bad, as it’s sometimes helpful to see the beam. Many years ago I worked with a copper-vapor laser doing high-speed imaging work. This was a laser with all of its power in two specific wavelengths, yellow and green. It was also powerful: 20 watts time-averaged output (with 2 kilowatts electrical input, it was a pretty good space heater as well). I started with a pair of safety goggles that blocked most visible wavelengths except for red. These let you see what you were doing in the lab, but you really couldn’t see the beam at all with these, either in the air or where it was hitting something and reflecting. After getting burned a couple of times by putting my hand in the beam, I got a pair of CV laser specific goggles. They blocked all of the green wavelength and most of the yellow wavelength, leaving enough for you to safely see (and avoid) the beam.

It’s a bit like ‘how significant is a headwind?’ It really depends on the particular conditions. My own guess is that scattering in what we’ll call clean air is largely dwarfed by the laser’s generation technique and optics.

It’s moved a bit past that. They’ve already started fielding systems on ships for operational testing with more of the current generation 150 kW systems slated to be fielded. So far it’s been two ships. The USS Ponce has the initial 30 kW system. The USS Portland had a 150 kW system installed last year.

The US Army isn’t far behind. They’ve already tested a 30 kW system mounted on a Stryker chassis for SHORAD (SHOrt Range Air Defense.) They are targeting having a fieldable 50 kW SHORAD system ready by Fiscal Year 2022.

So shining a laser pointer up in the air isn’t going to help at all if you’re lost in the woods I guess.

Depends on how close the search party is (and if they’ve ruined their night vision with big flashlights).

I have a little pen-sized laser pointer that makes an extremely visible beam in clear air (optimistically, it’s rated at 100 mW, but I doubt it’s really more than 50). Not hard to see pointed straight up even from a couple hundred meters away, but the beam is so narrow that I doubt you could see it from a kilometer unless the beam hit a cloud.

And to answer the OP, scattering from the atmosphere is enough to see the beam. It’s easy to tell the difference between it hitting dust vs. N2/O2 scattering–grains of dust going through the beam leaves a bright “spark”, while the scattering is just an even haze.

Probably somewhere in that ballpark (within an order of magnitude or two). Rough guess, my laser is around 1 cd/m[sup]2[/sup] surface brightness from the side. The sun is around 10[sup]9[/sup] cd/m[sup]2[/sup]. I’d suppose you’d need something at least 10x the sun’s brightness to do permanent damage. And my laser’s beam is quite thin, whereas something with this power density is going to have a much fatter beam. You get more brightness from having a higher depth, but lose some from the lateral spread. So, terawatt is probably not far from the truth.