The Rise of The Machines - Can AI Ever be Safe?

In this candid and insightful session from the Artificial Unintelligence Conference 2025, Anthony Aarons (Enterprise Risk Expert and former Global Risk Director at HSBC) takes on one of the most critical questions in technology today: Can AI ever truly be safe?

Anthony argues that the real risk in AI doesn’t come from the machines — it comes from us. Through a lens of enterprise governance and operational risk, he explores how leadership decisions, misaligned incentives, and the absence of accountability create the conditions for unsafe AI systems. The talk offers a pragmatic yet hopeful vision for how organizations can build adaptive risk frameworks that evolve alongside AI itself.

Key themes include:

Why AI safety is a human governance problem, not a technical one

The widening gap between AI innovation and risk maturity

Rethinking operational risk frameworks for self-learning systems

Why “human in the loop” must be more than a compliance phrase

How integrity and leadership define the path to safe AI