#035 - The AI Safety Crisis: Why We Are Not Ready with Dr Paul Dongha

Governing the Machine, out 23rd October on Amazon and other retailers. Pre-order now: https://amzn.eu/d/9PSBRAtGoverning the Machine is the essential guide to harnessing the transformative potential of AI, while navigating and mitigating its inherent risks. Authors: Ray Eitel-Porter, Dr Paul Dongha, and Miriam Vogel.This episode dives deep into AI safety, ethics, and risk governance with Dr. Paul Dongha, co-author of Governing the Machine. The conversation traces the evolution of AI from early agentic systems to today’s large language models, highlighting the risks of unchecked adoption: bias, unpredictability, data leaks, prompt injection attacks, and deepfakes. The conversation reflects upon the societal parallels with social media, the dangers of automation bias, and the urgent need for regulation, corporate responsibility, and critical thinking. Ultimately, the discussion frames AI as a powerful but unfinished technology that must be carefully navigated to ensure it benefits humanity rather than undermines it.Dr Paul Dongha is a pioneering figure in the transformational potential of AI and stands at the forefront of integrating AI within Financial Services, championing an approach that harmonizes technological advancement with corporate integrity and the Responsible use of AI. With a career spanning over 30yrs, Paul has cultivated a reputation for visionary leadership, guiding organisations in seizing AI opportunities, whilst maintaining a steadfast commitment to ethical principles.Paul is the first Bank-wide Head of Responsible AI and AI Strategy at the UKs largest bank, leading a dedicated team of ethics practitioners, and pioneering Responsible AI in the UK.