The higher-energy precursor of the AdS/CFT correspondence

To arrive at the AdS/CFT correspondence, Maldacena (building upon pre-existing work, most notably that of Klebanov et al.) assumed that already at the string theory level there exists a duality between the two alternative descriptions of D-brane physics, and then proceeded to take a low-energy limit, which in particular decouples the branes from the bulk. In this paper, aimed primarily at non-string-theorists, we review some of the issues one encounters when trying to formulate a precise duality statement at the level of string theory (prior to taking any limits). We also survey some of the results obtained in the study of the correspondence in the intermediate-energy regime where excited string modes are negligible but the branes are still coupled to the bulk. In this simplified context, we explain in particular how symmetries have been used to determine the form of the D3-brane effective action relevant to the duality, go over the derivation of a recipe to compute correlation functions in this theory, and discuss some of the questions that remain open.