Small and medium-sized enterprises (SMEs) increasingly face pressure to adopt artificial intelligence (AI) as a driver of competitiveness and innovation. While AI-as-a-Service (AIaaS) offerings have lowered technical and financial entry barriers, many SMEs still struggle to translate their general aspiration to “use AI” into concrete, value-creating applications. This paper addresses this challenge by designing and evaluating an interactive AIaaS toolbox that empowers SMEs to “do AI themselves”. Following a design science research approach, we derive design knowledge comprising design requirements, principles, and features grounded in practitioner insights from multiple SMEs of different industries. An instantiated artifact aims to support decision-makers in identifying and aligning AI use cases with organizational goals while fostering literacy, responsibility, strategic guidance, business model development, and regulatory awareness. We evaluated the design knowledge and prototype through semi-structured practitioner interviews, which confirmed the relevance of all proposed principles and underscored the need for accessible guidance mechanisms. Participants particularly appreciated the modules on archetypal use cases and regulatory evaluation. By providing empirically validated design knowledge for socio-technical adoption support systems, this study advances both theoretical understanding of SME-oriented AI adoption and practical approaches for implementing sustainable AIaaS initiatives.