Saleh Seyedzadeh

In the rush to adopt AI and data-driven systems, we often focus on algorithms, pipelines, and platforms — but overlook the critical foundations of trust and transparency. Without them, even the most sophisticated models risk being sidelined or rejected by the very people they are meant to serve. This session will explore how to build confidence in data systems by reducing bias, embedding ethical safeguards, and ensuring communication bridges rather than widening the gap between technical and non-technical stakeholders. Drawing on lessons from The Data Lab’s national programmes — from the Smart Manufacturing Data Hub (SMDH), where SMEs demanded transparent, usable dashboards, to the Participatory Harms and AI Workbench Monitoring (PHAWM) project, which monitors AI harms and hallucinations — we will highlight practical ways to design systems that people can genuinely trust. Attendees will leave with actionable frameworks and examples for embedding ethics, transparency, and human-centric thinking into their own data projects.