By Yashoda Fezah, General Manager, Compliance Administration and Support Services (CASS)
The recently released 2024 Quest State of Data Intelligence Report highlights that 2025 will be a transformative year for data intelligence. A significant finding is that data governance has evolved rapidly – what was once a legal compliance matter has now become an integral part of AI strategies.
Even as the top drivers for data governance continue to be data quality improvements at 42%, with security and analytics tied in second place at 40%, AI has made firm strides with 34% respondents opting for it. This clearly emphasises the increasing focus of AI’s role in data governance and data readiness.
Simply put, AI includes all forms of computational techniques that mimic human abilities. Gartner explains AI as “applying advanced analysis and logic-based techniques, including machine learning, to interpret events, support and automate decisions, and take actions.”
Meanwhile, the Data Management Association International (DAMA) defines Data Governance as “the exercise of authority, control and shared decision-making (planning, monitoring and enforcement) over the management of data assets.“
Globally, AI is reshaping industries and businesses, requiring organisations to remain vigilant and accord higher priority to data governance. As the impact of AI in organisational strategies and decision-making deepens, organisations must reevaluate their data security, privacy, and governance strategies relating to AI apps.
In Mauritius, regulators have been leading from the front in AI advancements. In February 2023, the Financial Services Commission launched an AI-powered Due Diligence Platform. This followed their issuance of the Robotic and AI Enabled Advisory Services Rules in June 2021 for financial services firms to broaden their offerings in a regulated space.
Understanding the role of AI in data governance and the trends that are shaping it is key for compliance professionals. Let us take a closer look at the main trends:
#1 The importance of safeguarding sensitive data
As AI evolves rapidly, organisations are moving beyond experimentation to tap into sensitive data for training AI models. Even as this opens up new business avenues with increased efficiencies, concern about sensitive data leakage is a natural fallout. Data that would have traditionally been locked in silos is now exposed to AI use cases.
Unlike securing data in standard databases, the security of data inputs and outputs from AI applications will make fresh demands on organisations. Here, we believe that data privacy professionals will play a key role in supporting companies to adapt their data security and governance frameworks to the new architecture.
# 2 A unified data security and governance strategy is key
With organisations relying on diverse AI tools, the need for a unified approach to data governance has never been greater. A unified data security and governance strategy is key to support AI-led transformation. Here, data leaders must apply controls to sensitive data and AI in a consistent and scalable manner.
A comprehensive data security approach requires an end-to-end lifecycle lens as follows:
– Find, classify, and tag where sensitive data might be located in your data estate.
– Next, secure your data access; and
– Finally, audit and monitor your data security posture.
# 3 An open and collaborative approach is a must
With AI being a nascent technology and companies continuously finding new ways to use AI, the future holds unbridled promise! Technology leaders are particularly excited about the convergence of AI with other technologies such as quantum, robotics, biotechnology, and neurotechnology.
However, as AI overlaps with other innovative technologies, an open technology architecture and collaboration with diverse stakeholders will be critical. As AI transforms industries, AI governance will be a key enabler in unlocking the full value of this technology.
#4 AI professionals are needed
Scaling AI also requires skilled individuals who are also able to offer support in specialised areas such as incident management, conformity assessments, transparency reporting and technical documentation. Skilled professionals will also be needed to meet the growing demand for effective regulatory compliance and self-governance.
Such professionals could emerge from the ranks of existing technology leaders or new start-ups focused almost exclusively on AI governance solutions such as AI inventory management, policy management and reporting.
#5 AI requires governance
While AI is being applied to data governance for greater efficiency, AI itself must be governed effectively. Data governance should focus on defining business cases, curating data sets taken for training, supervising data sets for production, and controlling models’ governance. As AI equates innovation, good governance for AI should balance the need to innovate with the risk and impact on data security of AI use cases. This requires organisations to:
– Take into account stakeholder concerns across a broad spectrum;
– Provide clear and effective guidance for practitioners;
– Integrate AI governance tools within the existing governance framework, and
– Remain flexible and agile to adapt.
The way forward
As we highlighted in a recent post on our LinkedIn page, the UN and OECD have announced a significant enhancement in their collaboration on global AI governance at the Summit of the Future in New York. This enhanced collaboration shall leverage the OECD’s AI Policy Observatory and the UN’s global network to create a more cohesive policy ecosystem for the advancing AI space.
In Mauritius, we celebrate it as a major step towards coordinated AI governance across borders! Against this milestone in fostering responsible AI development that benefits everyone, we look to the future with interest and excitement for nuanced AI use cases that balance innovation with risk.
