Carbon dioxide (CO
2
) has long been considered, at best, a waste by-product of metabolism, and at worst, a toxic molecule with serious health consequences if physiological concentration is dysregulated. However, clinical observations have revealed that ‘permissive’ hypercapnia, the deliberate allowance of respiratory produced CO
2
to remain in the patient, can have anti-inflammatory effects that may be beneficial in certain circumstances. In parallel, studies at the cell level have demonstrated the profound effect of CO
2
on multiple diverse signalling pathways, be it the effect from CO
2
itself specifically or from the associated acidosis it generates. At the whole organism level, it now appears likely that there are many biological sensing systems designed to respond to CO
2
concentration and tailor respiratory and other responses to atmospheric or local levels. Animal models have been widely employed to study the changes in CO
2
levels in various disease states and also to what extent permissive or even directly delivered CO
2
can affect patient outcome. These findings have been advanced to the bedside at the same time that further clinical observations have been elucidated at the cell and animal level. Here we present a synopsis of the current understanding of how CO
2
affects mammalian biological systems, with a particular emphasis on inflammatory pathways and diseases such as lung specific or systemic sepsis. We also explore some future directions and possibilities, such as direct control of blood CO
2
levels, that could lead to improved clinical care in the future.