Magnetic resonance imaging (MRI) has added a new dimension to the study of osteoarthritis, a long-known degenerative joint disease with limited therapeutic options. It has advanced our understanding of joint pathophysiology and identifying that osteoarthritis as a simple ‘wear and tear’ process of the articular cartilage has indeed become a thing of the past. Recent work has focused on the study and validation of MRI scoring/quantification systems, as well as the identification of MRI predictors of symptoms/disease progression. The latter may serve to identify patients at greater risk for osteoarthritis disease progression to be enrolled in clinical trials. Like all imaging tools, MRI use has its associated problems. Structural changes seen in patients with osteoarthritis are often seen in asymptomatic subjects and this makes an MRI definition of osteoarthritis less straightforward. The ability to pick up multiple structural abnormalities simultaneously and high sensitivity in delineating structural changes can makes interpretation of true pathology more complicated. Although there has been much progress in the field of MRI in osteoarthritis, there remain many clinical/technical issues that need to be addressed. Until more data are obtained from clinical trials, the question of whether MRI is useful in therapeutics intervention in osteoarthritis remains unanswered.