We use an empirical model of global climate (Canty et al., ACP, 2013 <http://www.atmos-chem-phys.net/13/3997/2013/acp-13-3997-2013.html>) to quantify the role of human activity on global mean surface temperature (GMST). The model determines values of the climate feedback and the ocean eddy diffusion coefficient that are consistent with both the observed global, monthly GMST anomaly and the rise in ocean heat content, upon consideration of effects on climate of a variety of anthropogenic (i.e., greenhouse gases and tropospheric aerosols) and natural (ENSO, major volcanoes, solar irradiance, thermohaline circulation) factors. This presentation will focus first on a comparison of the attributable anthropogenic warming rate (AAWR) inferred from GMST for the past three decades using our model to AAWR found by the full-scale climate models that participated in CMIP5, the basis for global warming projections used in IPCC (2013). We’ll show that most of the CMIP5 models warmer faster than the actual climate system. We will then touch upon the global warming hiatus, a period of purported decline in the rate of global warming between 1998 and 2010. Finally, we’ll show projections of global warming using our approach in probabilistic terms, for which the largest uncertainty centers around cooling due to tropospheric aerosols. We will conclude by showing how the rise in global mean surface temperature can be kept below 2ºC warming relative to a pre-industrial baseline, which is the goal of the recent UNFCCC-COP21 meeting held in Paris.