Abigail Tate
Business Development Manager – IBM Software | TD SYNNEX UK&I
Curating trust with IBM watsonx.governance, IBM watsonx.ai, IBM Granite, IBM wastonx.data, and IBM Guardium AI Security
How do you create the human need for trust in the world of Artificial Intelligence? As AI has evolved from experimentation to enterprise infrastructure, trust has become a business imperative. It’s a topic I’ve enjoyed exploring as I get to know IBM watsonx.governance and share with you a surprising case study!
The rapid rise in risk
Alongside the emergence of generative and agentic systems, the risks of algorithmic bias, security exposure and regulatory non-compliance have accelerated. For data leaders, this means it’s no longer a case of whether AI ethics and governance matter; it’s a strategic imperative.
Today, AI governance is no longer a defensive move to avoid reputational damage; it’s a strategic priority to scale safely, build trust and unlock sustainable value. Here’s why.
1: Governance enables enterprise AI at scale
As AI initiatives gain traction beyond isolated pilots, complexity increases. This leads to multiple models, diverse data sources, complex third-party APIs, and evolving regulatory requirements. Without a strong governance framework in place, AI infrastructure becomes confusing and difficult to manage.
IBM watsonx.governance helps data leaders manage AI risk and compliance across the lifecycle of models, applications, and agents. These governance capabilities include monitoring model behaviour, detecting bias, tracking drift, and providing explanations around automated decisions. Tools such as IBM watsonx.ai and foundation models like IBM Granite give organisations the ability to build and deploy AI applications while embedding governance and transparency into the development process.
2: You’re (still) only as good as your data
We all know that bad data = bad business decisions. The challenge with AI is that in the case of bias, or poor governance, AI will amplify issues rather than resolve them.
As a result, modern data architectures need to be built around governed data platforms. IBM watsonx.data is a great example as it provides an open lakehouse architecture to support trusted, AI-ready data across hybrid environments. By enabling organisations to manage lineage, access controls and data quality, watsonx.data ensures that AI systems are trained and deployed on reliable information.
3: Trust is a competitive advantage
The World Economic Forum recently published an Emerging Technologies article predicting that ‘Trust’ will become one of the most valuable currencies in the AI economy and that business stakeholders – customers, regulators, employees – are looking for organisations to be open and accountable in how they use AI.
At the same time, the AI regulatory landscape is tightening up. Governments are introducing frameworks to address algorithmic transparency, bias mitigation, and accountability for automated decision-making. With corporate and public sector bodies pushing to ensure that AI solutions are compliant, transparent, and aligned with regulatory requirements, the need to prove that solutions are auditable and trusted is vital.
To help with this, tools such as IBM Guardium AI Security provide key capabilities to help organisations with the task of identifying shadow AI deployments, detecting vulnerabilities, and protecting sensitive data used in AI models. When combined with governance platforms, these capabilities create a comprehensive approach to managing AI risk that satisfies internal and external stakeholders – and avoids the risk of costly retrofits and operational disruption.
AI Governance in Action: The Ultimate Fighting Championship (UFC)
And here’s that case study! For most people, the UFC conjures up visceral images of elite fighters, but as an organisation, it’s leading the pack in AI governance. Using generative AI to create real-time insights, predictions, and narratives during live events, the UFC enhances the fan experience by providing deeper context around fighters’ performance and strategy.
However, the UFC recognised early that automated narratives can introduce risk. Would AI outputs misrepresent fighters? Could hidden bias affect predictions? Would this impact the organisation’s credibility with fans?
To address these questions, the UFC uses IBM watsonx.governance to review AI-generated insights, detect bias, validate outputs, and provide transparency around how predictions are produced.
For example, when the system predicts the outcome of a fight, governance tools explain which performance metrics were used. This level of transparency ensures AI insights are engaging and trustworthy.
As a result, the UFC increased the number of AI-generated insights delivered during events by three times while reducing query generation time by 40 percent – enhancing fan engagement without sacrificing integrity.
AI Governance as a Leadership Imperative
Whilst I won’t be stepping into The Octagon any time soon, the developments made by the UFC feel like a natural evolution of a lesson I learned early in my career as a database administrator: technology may change, but trust in data remains the foundation of every digital system.
This article was written by Abigail Tate, Business Development Manager for IBM Software at TD SYNNEX. For more information, contact Abi by emailing her at abigail.tate@tdysynnex.com or by completing the form below.
