For AI researchers and data scientists, developing a high-performing AI solution is often the finish line—but in a critical environment like Transurban, it is only the starting point. This talk explores how our day-to-day work at Transurban has evolved through the adoption of AI and cloud platforms, underpinned by strict governance to maintain organisational confidentiality. We will first examine the real-world friction experienced when transitioning models from development to deployment. Our daily workflows now require navigating highly governed cloud environments, securing access to sensitive data, and adapting our analytical approaches to address valid privacy and safety concerns.

I will share high-level insights into how we overcome these roadblocks to safely scale "Everyday AI". We will explore how enterprise-wide Gemini adoption and our multi-cloud ecosystem empower analytics workflows instead of restricting them. By unpacking dynamic risk frameworks and automated governance from a practitioner's perspective, this session will demonstrate how to bridge the gap between theoretical data science and delivering resilient, governed enterprise AI.