As artificial intelligence moves from experimentation into everyday business operations, the question is no longer whether organisations should adopt AI, but how they can do so responsibly. The challenge is not simply deploying new capabilities – it’s understanding when AI can be trusted, where it can fail and what safeguards are needed before it is put into practice.

In this excerpt from episode 232 of Scouting For Growth, Sabine VanderLinden sits down with Reggie Townsend, Vice President of AI Ethics, Governance and Social Impact at SAS, to explore the growing AI trust deficit and why organisations need to rethink what it means to trust an AI system. Drawing on his work around responsible and trustworthy AI, Reggie explains why trust should not be treated as binary, and why the level of confidence placed in an AI system needs to reflect the consequences of getting it wrong.

Together, they explore the tension between underusing reliable AI because confidence is too low and over-relying on probabilistic systems because confidence is too high. Reggie also explains why the distinction between deterministic and probabilistic AI becomes particularly important in high-risk sectors such as financial services, healthcare and insurance.

Sabine and Reggie discuss:

• Why responsible AI must move beyond asking whether organisations should use AI, towards understanding for what purpose, to what end and for whom AI might fail.
• What the AI trust deficit looks like in practice, from organisations underusing reliable systems to employees placing too much confidence in generative AI outputs.
• Why trust should be viewed as a range rather than a binary, with different AI systems appropriate for different levels of risk.
• The crucial distinction between deterministic and probabilistic AI, and why high-risk decisions require a different level of certainty and control.
• How governance, explainability and safeguards can help organisations close the gap between claiming to trust AI and actually making AI systems demonstrably trustworthy.

As AI becomes increasingly embedded in business and society, building trust will require more than confidence in the technology itself. Organisations will need to understand its limitations, match systems to the risks involved and create the governance needed to use AI responsibly. This conversation explores what that looks like – and why getting the balance right will be critical to unlocking AI’s full potential.

Listen to the full episode now on Apple, Spotify and Podbean.

To browse the full catalogue of Scouting for Growth episodes, click here.