Where Responsible AI meets Reality: Practitioner Perspectives on Enablers for shifting Organizational Practices

Organizational culture and structure matter as much as technical solutions: The paper finds that responsible AI initiatives fail or succeed based on how companies are organized and their internal culture, not just the availability of technical tools—suggesting that companies need systemic changes beyond hiring ethics teams.

Practitioners face real ethical tensions with no clear resolution: Industry workers report genuine conflicts between responsible AI goals and business pressures, indicating that responsible AI isn't simply a matter of "doing the right thing" but involves navigating competing organizational demands.

Current company structures actively hinder responsible AI work: The research identifies specific organizational barriers that prevent responsible AI initiatives from being effective in practice, though the paper indicates these barriers can be addressed through intentional structural redesign.

A roadmap exists for transitioning to better practices: Rather than presenting responsible AI as an unsolvable problem, the paper maps concrete elements needed to move from current practices to more effective future processes, offering practitioners actionable guidance for organizational change.

Real-world implementation differs significantly from responsible AI theory: The gap between what responsible AI frameworks recommend and what actually happens in production systems reveals that practitioners need support addressing practical constraints, not just ethical principles.