?
Risks and the image of the future in the study of AI technologies prospects
This study addresses the systemic identification and categorization of risks associated with AI development, arising from tensions between technological evolution and institutional, infrastructural, and economic contexts. Drawing on a constructionist methodology, we interpret technological risks as constitutive elements of expert communities' images of the future. Through in-depth interviews with 100 AI experts, proportionally representing corporate, research, and regulatory segments of the AI innovation ecosystem we identify and code 16 distinct risks. The study operationalizes the Polak matrix to map technological risks along two dimensions: essence (the perception of AI development processes) and influence (the role of professional agency). Applying the matrix, the risks are categorized into four groups: Disappointing AI, AI Winter, Biased AI, and Absolute AI, each exhibiting distinct managerial properties and contributing to the scenario space of AI development. The proposed framework reveals the ambivalent nature of technological risk, distinguishing between causal ambivalence (process vs. environmental attribution) and evaluative ambivalence (varying threat intensity estimates). The authors further demonstrate how the resulting risk matrix can improve strategic planning by projecting risk data onto a modified SWOT framework, offering actionable pathways to navigate the socio-technical complexity of AI governance.