This study is concerned with the problem of adaptive sliding mode control (SMC) for uncertain stochastic singular systems. A novel integral-type sliding surface function is first introduced based on a particular observer design, which incorporates both the state estimations and the outputs to achieve prescribed specifications. The analysis of mean-square asymptotic admissibility of the underlying sliding mode dynamics with <alternatives>$\mathscr {H}_{\infty }$<mml:math overflow="scroll"><mml:msub><mml:mrow><mml:mi mathvariant="script">H</mml:mi></mml:mrow><mml:mrow><mml:mi mathvariant="normal">∞</mml:mi></mml:mrow></mml:msub></mml:math><inline-graphic xlink:href="IET-CTA.2016.0911.IM2.gif" /></alternatives> disturbance attenuation level for the closed-loop systems is performed to exploit a new condition via linear matrix inequality technique. Then, the reachability of the predesigned sliding surface is ensured within finite-time almost surely by utilising a novel adaptive SMC law. Two examples are provided to demonstrate the validity and potential of the proposed method.