Abstract This study develops an activity‐based inversion framework to quantify sectoral carbon dioxide (CO2) emissions using observations of co‐emitted air pollutants. It leverages distinct sectoral co‐emission profiles of nitrogen oxides (NOx), sulfur dioxide (SO2), and carbon monoxide (CO) to attribute sources and optimize sectoral activity‐rate scaling factors, which are applied to bottom‐up CO2 emissions to estimate posterior emissions. Results show that nitrogen dioxide (NO2) informs energy and transportation adjustments, while SO2 and CO constrain industrial and residential adjustments. Pseudo‐observation tests with known solutions show that the posterior reduces normalized mean square error (NMSE) by 26%–68% and normalized mean bias (NMB) by 10%–63% relative to the prior when evaluated against true CO2 emissions. Posterior simulations show improved agreement with Greenhouse Gases Observing Satellite (GOSAT) CO2 dry column mixing ratios, but gains are limited where CO model‐observation mismatches drive large residential emission adjustments, highlighting the need to reduce and better characterize CO‐related errors.