Obesity and type 2 diabetes are associated strongly with NAFLD. It is not clear if one of these conditions causes the others, or if all are consequences of another process. Although NAFLD is known to occur in overly lean individuals, which indicates that excessive adiposity is not required for the development of NAFLD, the severities of insulin resistance and NAFLD tend to parallel each other, and the greatest prevalence of type 2 diabetes occurs in patients with NAFLD and cirrhosis. This observation suggests that insulin resistance and NAFLD may be related pathogenically. Experiments in mice demonstrate that insulin resistance and NAFLD result from a chronic inflammatory state that is characterized by increased levels of TNF alpha. The mechanisms that drive this chronic inflammation are unknown but might involve the oxidative stress that develops during fatty acid metabolism or when products from intestinal bacteria escape into the mesenteric blood to trigger a sustained hepatic inflammatory cytokine response in genetically susceptible individuals, promoting a positive feedback loop that reinforces insulin resistance and inflammation. This hypothesis is supported by some animal and human studies; however, more research is needed to evaluate this theory. Additional studies also are required to determine the benefits of treatments that interrupt this pathogenic cascade at various points. Preliminary work in animal and human studies suggests that diverse strategies that inhibit production of TNF alpha and improve insulin resistance also ameliorate NAFLD.