Avoid an AI governance blind spot!
Board members need to seriously consider artificial intelligence (AI) governance as AI becomes integral to business operations. Understanding its impact is key to fulfilling their duty of care, managing risks and guiding their organisations in using AI responsibly and ethically.
As I took on more governance roles, I recognised that boards play an increasingly critical role in overseeing AI, but many are unprepared for this responsibility. This led me to focus on how boards can proactively manage AI governance to avoid being left behind as AI becomes more prevalent.
Shadow AI, for example, is an alarming trend of unsanctioned generative AI use within an organisation. Employees may access AI tools to perform a range of tasks, including drafting copy or writing code. However, a recent Cyberhaven report, which surveyed 3 million employees, found that 73.8% of workplace ChatGPT usage occurs through public accounts. Generative AI tools can help drive innovation and agility, but they also come with significant risks. The 2023 Samsung incident is a stark reminder: employees used generative AI to analyse sensitive data, inadvertently leaking source code—a clear example of the dangers of using public AI tools without robust safeguards.
These headlines should serve as a wake-up call for boards to proactively tackle AI governance. The challenge is determining which AI uses to permit or restrict, balancing employee support with business security.
But where should boards start?
- Conduct an AI audit
Identify how AI is currently being used and address any Shadow AI risks. - Develop a comprehensive AI policy
Collaborate with management to develop a comprehensive AI policy, addressing ethical use, data privacy, security measures, and guidelines for AI procurement and development. - Understand AI risks
Utilise resources like the AI Risk Repository to help identify potential issues. - Foster a culture of responsible AI
Promote ethical AI practices, emphasising accountability, transparency and security. - Determine AI oversight
Determine where AI oversight will reside—whether with the full board, the Audit and Risk Committee, or another designated committee. - Consider an AI framework
Use frameworks like ISO/IEC 38507 AI Governance, NIST AI Risk Management Framework and Microsoft Responsible AI Standards to guide you. - Provide AI training for board members
Invest in training that has been designed specifically for boards.
By Karen Rolleston
Karen is a chartered member of the Institute of Directors and a professional director. If you would like to know more about AI board training, please email karen@kaw.net.nz
Keep updated with the AI Forum’s news and events here. If you’re not already a member, learn more and join us.
