> The only atmosphere being cleaned is that which passes through the machine. This is like saying that the only gasoline that is consumed is that which is burn

> The only atmosphere being cleaned is that which passes through the machine. This is like saying that the only gasoline that is consumed is that which is burned up in a car. 1) A car can still be useful without consuming all gasoline in the world. This machine could be economically viable given carbon credit structures as well, it doesn't have to scale to solve the problem on its own, it just has to be efficient enough to be worth building (I'm not claiming that it meets this criteria). 2) Clearly we could actually use up the majority of gasoline on earth, there is no reason why scaling something that much is impossible. > with currents and what-not, meaning this machine needs to be all over the globe to make meaningful impact. Are you really saying this is an obstacle-- that if we wanted to make 1 million of these machines we would be unable to distribute them, and that furthermore we would completely clean the co2 from the atmosphere over a single country before the currents and what-not would bring in more co2 for the machines to consume? If this is true, then why do we care about co2 emissions in other countries currently? > We don't know the emission cost of producing these machines - but it certainly isn't zero. We also don't know the emission cost of the energy being produced to power these things (which even with geo-thermal, is not zero either). I agree that we need to know this and the other factors and that they are larger than zero, but it is critically important to know the actual values, or at least a rough estimate. How are you so confident whether it would work or not without knowing them, or even giving an estimate and doing the napkin math?