Type 2 diabetes mellitus (T2DM) is characterized by an ∼60% deficit in β-cell mass, increased β-cell apoptosis, and islet amyloid derived from islet amyloid polypeptide (IAPP). Human IAPP (hIAPP) forms oligomers, leading to either amyloid fibrils or toxic oligomers in an aqueous solution in vitro. Either application of hIAPP on or overexpression of hIAPP in cells induces apoptosis. It remains controversial whether the fibrils or smaller toxic oligomers induce β-cell apoptosis. Rifampicin prevents hIAPP amyloid fibril formation and has been proposed as a potential target for prevention of T2DM. We examined the actions of rifampicin on hIAPP amyloid fibril and toxic oligomer formation as well as its ability to protect β-cells from either application of hIAPP or endogenous overexpression of hIAPP (transgenic rats and adenovirus-transduced β-cells). We report that rifampicin (Acocella G. Clin Pharmacokinet 3: 108–127, 1978) prevents hIAPP fibril formation, but not formation of toxic hIAPP oligomers (Bates G. Lancet 361: 1642–1644, 2003), and does not protect β-cells from apoptosis induced by either overexpression or application of hIAPP. These data emphasize that toxic hIAPP oligomers, rather than hIAPP fibrils, initiate β-cell apoptosis and that screening tools to identify inhibitors of amyloid fibril formation are likely to be less useful than those that identify inhibitors of toxic oligomer formation. Finally, rifampicin and related molecules do not appear to be useful as candidates for prevention of T2DM.