A typical UK office is used as an exemplar for office energy use in the UK. The effect of replacing existing boilers and air-conditioning systems with air-source heat pumps (ASHPs) is investigated for a "current" version of the office, with typical equipment/lighting usage, fabric and internal gains, and also for a "2030" office, where fabric is improved, equipment/lighting made more efficient and, as a result, internal gains reduced. The ASHPs, as a potential carbon-saving technology, performs slightly differently for the two office scenarios. Furthermore, after removing the boiler, it is found to be important whether electric hot water or gas hot water point-of-use heaters are adopted with the ASHPs (assuming that the existing boiler would not be used if the ASHPs is satisfying all space heating requirements). This can be the difference between ASHPs reducing and increasing the carbon emissions of the office. Finally, the carbon intensity of the grid has a large effect on the success of ASHPs technology. This is quantified through a sensitivity analysis, indicating the external conditions for which ASHPs might reduce CO2 emissions for office buildings. The results suggest that an ASHPs has the potential to reduce CO2 emissions for certain conditions, but should not be seen as a guaranteed low-carbon technology for all scenarios. As well as assessing the ASHPs as a carbon-saving technology, potential economic benefits are also estimated based on running costs and predicted reduction in energy bills. (C) 2008 Elsevier B.V. All rights reserved.