Improving AI Policy Through Multistakeholder Engagement
==Multistakeholder engagement can improve AI policy and system design by making governance more balanced, bringing wider expertise into the AI lifecycle, strengthening accountability, and ensuring that the perspective...
==Multistakeholder engagement can improve AI policy and system design by making governance more balanced, bringing wider expertise into the AI lifecycle, strengthening accountability, and ensuring that the perspectives of affected communities shape both policy and technical design.[:cite[2]{ln=1}][:cite[2]{ln=2}][:cite[1]{ln=4}][:cite[1]{ln=5}][:cite[1]{ln=6}]== How it improves AI policy == It creates a more balanced regulatory environment. == The report says the AI ecosystem is “cross sectoral, complex and diffuse,” and therefore effective governance requires “multistakeholder involvement” plus collaboration and coproduction to ensure balance in regulation.[:cite[2]{ln=1}][:cite[2]{ln=2}] == It broadens the knowledge used to make policy. == The report argues that effective AI governance must look beyond technical solutions and include social sciences and humanities knowledge.[:cite[1]{ln=4}] It specifically points to public participation, auditing, impact assessments, and GenAI assessments as methods that show why this broader knowledge base matters.[:cite[3]{ln=1}] It also recommends including social sciences and humanities expertise in standards, guidelines, AI assessment, research, development, and policy.[:cite[3]{ln=2}] == It supports iterative and adaptive governance. == The report says effective governance is enhanced by links to regulation, specificity, enforceability, monitoring, and “iteration and follow up.”[:cite[1]{ln=1}][:cite[1]{ln=3}] In its conclusion, it adds that ongoing stakeholder engagement and iterative regulation will be vital for adapting to emerging challenges.[:cite[4]{ln=3}] == It helps prevent power imbalances and democratic erosion. == The report’s key findings say effective governance needs “multistakeholder accountability frameworks” so AI does not reinforce existing power imbalances or erode democratic norms.[:cite[5]{ln=2}] How it improves AI system design == It improves outcomes during development, testing, and use. == The report states that when sociotechnical approaches are integrated into AI development, testing, and use feedback, positive outcomes significantly increase for impacted communities, users, and AI developers.[:cite[1]{ln=5}][:cite[1]{ln=6}] == It embeds human values across the full AI lifecycle. == The report says operationalising human values in AI requires broadening the disciplines involved across planning, design, development, and deployment, and building tools for effective cross disciplinary work.[:cite[6]{ln=1}] == It informs procurement and evaluation, not just abstract ethics. == The report recommends mandated use of sociotechnical research and evaluation methods in procurement processes, alongside broader expert inclusion in standards and assessment.[:cite[3]{ln=2}] Why participation by affected groups matters == It allows non experts and affected communities to shape both policy and models directly. == The report describes WeBuildAI as a participatory framework that involves lay users directly in the design of algorithmic governance policy and allows non expert communities to be directly involved in the design of the AI model itself.[:cite[7]{ln=1}][:cite[7]{ln=2}] == It increases diversity among those who build, regulate, and deploy AI. == The same passage says emerging AI collaboratives focus on enhancing diversity in the communities that build, regulate, and deploy AI.[:cite[7]{ln=3}][:cite[7]{ln=4}] == It surfaces perspectives that standard expert led processes may miss. == In the report’s example of children’s engagement, children developed calls to action that included inclusion, avoiding negative impacts, involving children in AI development, protecting children’s rights, opt in consent to data use, and embedding AI literacy in the curriculum.[:cite[8]{ln=1}][:cite[8]{ln=2}] The report says these ideas came directly from children and pushed toward a more normative climate for AI development.[:cite[8]{ln=3}] It also notes that these calls were presented to decision makers and intended to ensure children’s voices are considered in AI decisions.[:cite[9]{ln=1}][:cite[9]{ln=2}][:cite[9]{ln=3}] Wider benefits == It builds trust, awareness, and internal change. == The report says multistakeholder collaboration is more likely to produce trust, workforce transformation, accountability mechanisms, and organisational self governance.[:cite[2]{ln=3}][:cite[2]{ln=4}] It also says such approaches can support public awareness and education.[:cite[2]{ln=5}] Bottom line ==In the report’s account, multistakeholder engagement improves AI policy and design because it connects governance to the real social contexts in which AI operates, brings affected people and non technical expertise into decisions, and creates stronger accountability and legitimacy than narrow technical or industry only approaches can provide.[:cite[10]{ln=7}][:cite[2]{ln=2}][:cite[1]{ln=4}][:cite[1]{ln=5}][:cite[5...