What I was suggesting with the cost vs benefit comment was that there are better ways to spend the money for the same or better power outcome.
The CO2 system has much in common with a N2O system in terms of the parts it uses. What is different between them will cost a (broadly) similar amount of money. N2O is more expensive to buy than CO2 but the performance gain on it is vastly more than anything possible out of the CO2 based system. Legality is another question.
Chilling the air intake cannot save any money, only increase the maximum available power output. Attempting to cool the intake air at part throttle using the A/C system is self defeating. The A/C system has to work harder to pull the extra heat out of the intake air, hence PINKY355's comment about perpetual motion. At higher loads, it's easier just to open the throttle more if more power is needed. Cold intake air is not necessarily what is needed at part throttle anyway.
At wide open throttle we already covered that the A/C system is turned off. Any extra power can only be made available by creating a cold heat sink when the A/C is working (at part throttle) and then cooling the intake air with that. That is why I was suggesting a practical way to do that might be with a (chilled) water injection system.
The rate of heat transfer due to conduction is a function of temperature and surface area (and a co-efficient for the materials used) so the CO2 bulb will work better than a car A/C because the boiling CO2 will be much colder than the A/C can achieve, even if the surface area is smaller. (The sublimation point - it doesn't boil at atmospheric pressure - of CO2 is -78.5C, a car A/C more like 0-5C.) The surface area even of the full length of an intake duct compared to that of a proper heat exchanger is small though.