Answers are grounded in this paper and cite their sources.
Software and information technology (IT) projects depend on external vendors, and their outcomes depend on how vendors are selected and on how the client–vendor relationship is governed after the contract. Artificial intelligence (AI) is expected to reduce the information asymmetries behind these problems, yet studies are dispersed across software engineering, information systems and operations research, and it remains unclear which acquisition problems AI addresses, from whose perspective and with what evidence. This systematic review, designed according to Tranfield et al. and reported following PRISMA 2020, combined Web of Science and Scopus searches with five rounds of snowballing and included 105 studies. In the dyadic client–vendor cycle, AI application is concentrated before the contract, in vendor and product selection (17 studies), proposals and bidding (18) and contract analysis (18), as well as risk in outsourced projects (16); relationship governance after the contract accounts for six studies. The vendor side holds the strongest evidence, and 28 of the 75 dyadic studies come from two large IT service providers. Software crowdsourcing platforms form the largest axis (30 studies). AI has thus supported formal governance and ex ante transaction costs, while post-contractual monitoring, relational governance and the client’s perspective remain open problems.
No takes yet. Share an insight, caveat, or question.
Bergamini et al. (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: