Agentic Artificial Intelligence and Machine Learning for Enterprise-Wide Intelligent Automation and Digital Transformation in Saudi Arabia under Vision 2030

Enterprise automation is moving on from simple software rules to systems that can take in a request, work out what tools to use, co-ordinate a range of specialist agents and then make decisions based on the feedback they receive. That said, greater technical autonomy doesn't in itself lead to organisational change. This integrative review looks at how agentic AI, machine learning and the existing forms of process automation can be brought together to enable enterprise-wide change in Saudi Arabia as part of Vision 2030. It draws on 30 DOI identifiable publications from 2020–2025, covering the technological, organisational and Saudi specific evidence. A concept-based analytical approach is used to separate out what is actually demonstrated from what is merely proposed in terms of architecture and to assess the degree to which the findings can be applied to regulated enterprise workflows. Six interrelated dimensions are identified: process suitability, intelligent orchestration, data and infrastructure, human capability, risk governance and value measurement. The proposed architecture for accountable automation is one in which decision rights, evidence provenance and exception handling are all situated around a bounded agent execution loop. The Saudi case studies do suggest that management buy-in and digital maturity are important, but provide very little in the way of direct evidence on the scale of agentic deployment. The review therefore does not accept that agent autonomy, economic benefit and public legitimacy are equivalent. It does provide a context sensitive design framework and a testable research agenda for enterprises that are looking for real transformation that is reliable, interoperable and inclusive rather than just a series of isolated proofs of concept. Longitudinal, sector specific evaluation is required before any claim can be made about causal national outcomes.