Maybe I misunderstand how CO2 lasers work. But my understanding is that a high-powered CO2 laser can be run using a gas supply from a tank/regulator. As mentioned elsewhere, this is typically only done in industrial settings, but some home laser cutting enthusiasts want more power. The tank contains some mixture of gases, primarily CO2, but containing other gases, likely some CO. If the tank/regulator to laser tube connection is leaking in a closed room, then the CO concentration in the air will increase.
This is speculation on my part based on my assumptions. If my assumptions are incorrect that's fine.
Yes, but those would feed CO2 into the laser tube, which I thought you just excluded ("I never said anything about the laser tube.").
As others have said, normal laser cutters have sealed tubes, although its of course possible they had some DIY rig for whatever reason. MIT students building their own CO2 laser just for fun doesn't seem all that odd, using it for long time in a cutter maybe more so. I don't know if a leak in a CO2 system would lead to CO poisoning or no. CO produced by the material cut seems more likely to me.
Sorry, I guess it is a bit unclear: I assumed the leak preceded the laser tube proper. When you hook a tank/regulator up to an evacuated tube, there can be leaks "before" the laser tube (I spent a month building vacuum chambers, and most of my leaks were in the connections between the vacuum pump and the chamber, not the chamber itself).
however, after thinking some more, I agree that CO from material cut is more likely.
The "gas supply" you're referring to is most likely an "air-assist". Laser cutting works better when you have a flow of air (just air), blowing the smoke out of the way.
Many folks use small compressors for this purpose.
This is speculation on my part based on my assumptions. If my assumptions are incorrect that's fine.