As demographics in developed nations shift towards an aging population, neurodegenerative pathologies, especially dementias such as Alzheimer’s disease, pose one of the largest challenges to the modern health care system. Since there is yet no cure for dementia, there is great pressure to discover potential therapeutics for these diseases. One popular candidate is curcumin or diferuloylmethane, a polyphenolic compound that is the main curcuminoid found in Curcuma longa (family Zingiberaceae). In recent years, curcumin has been reported to possess anti-amyloidogenic, antiinflammatory, anti-oxidative, and metal chelating properties that may result in potential neuroprotective effects. Particularly, the hydrophobicity of the curcumin molecule hints at the possibility of blood-brain barrier penetration and accumulation in the brain. However, curcumin exhibits extremely low bioavailability, mainly due to its poor aqueous solubility, poor stability in solution, and rapid intestinal first-pass and hepatic metabolism. Despite the many efforts that are currently being made to improve the bioavailability of curcumin, brain concentration of curcumin remains low. Furthermore, although many have reported that curcumin possesses a relatively low toxicity profile, curcumin applied at high doses, which is not uncommon practice in many in vivo and clinical studies, may present certain dangers that in our opinion have not been addressed sufficiently. Herein, the neuroprotective potential of curcumin, with emphasis on Alzheimer’s disease, as well as its limitations will be discussed in detail.