The retirement industry has been marked by innovation since its inception. But no change happens faster than the AI revolution. This transformation will lead to better member outcomes, more knowledgeable members, more knowledgeable trustees, and eventually a reduction in costs due to fewer man-hours needed to service and educate members.
This is not to say there aren’t challenges at present. Member and trustee education on the safe use of AI, as well as an awareness of its risks, is essential. But can the industry really afford to be so cautious that it doesn’t move forward? Can trustees afford to refuse AI simply because they don’t know how to use it, or because they worry it might occasionally give an inaccurate response?
Where does the responsibility lie? Should individual trustees upskill themselves on AI use? Should service providers ensure trustees are equipped to leverage AI for more informed decision-making?
The answer is both: it’s a shared responsibility that will determine how quickly our industry can harness AI’s transformative potential.
How AI can assist board members in making better decisions
AI can become a thought partner for board members in ways we’ve never seen before. It can compare financial statements with remarkable speed and accuracy. It possesses extensive knowledge of the South African retirement fund landscape, encompassing laws, accounting practices, the latest FSCA Joint Standards, and regulatory changes.
Some of the ways in which AI can assist trustees include:
- Explaining and comparing outcomes of different investment strategies and portfolio options;
- Drafting clearer, more effective member communication;
- Explaining and summarising rules and rule amendments on the fly;
- Analysing financial statements and extracting the most critical information; and
- Doing complex calculations in seconds rather than hours.
But perhaps most importantly, AI can ask trustees the right questions. Board members and boards aren’t often challenged with questions about their decisions or asked to consider alternative perspectives when they make important choices. AI can now prompt trustees with questions about crucial issues they might not have considered, which ensures more thorough decision-making processes.
‘But perhaps most importantly, AI can ask trustees the right questions.’
Should boards simply rely on AI-generated recommendations or proposals?
No, AI should not make decisions for boards. It serves as a powerful tool, delivering higher-quality information faster. It will help trustees access information faster, perform calculations more efficiently, and assist the board to become less dependent on certain service providers for basic analysis and guidance.
Ultimately, the board retains full responsibility for decisions.
The reality of AI limitations
It’s crucial to understand that an AI model thinks for itself. Even when guardrails are built into an AI system, it will still make decisions based on all the material it has been trained on. This raises an important question: Will we see independent AI trustees running funds and sitting on boards in the future?
In my opinion, this is highly unlikely. AI is designed to analyse, guide, prompt, calculate, and assist with better decision making, not to replace human judgement entirely. Trustees cannot and should not depend solely on AI decisions and proposals. The human element, with all its experience, intuition, and accountability, remains irreplaceable.
‘The human element, with all its experience, intuition, and accountability, remains irreplaceable.’
The path forward
The AI revolution in retirement funds isn’t about replacing people; it’s about augmenting human capability. When trustees can quickly analyse financial statements, understand complex legislation, and craft member communication that truly resonates, they’re better equipped to serve their members’ interests.
This transformation requires a mindset shift. We need trustees who are willing to learn, service providers who prioritise education, and an industry that embraces innovation while maintaining appropriate safeguards.
The retirement industry stands at a crossroads. We can either embrace this technology thoughtfully and strategically, or we can allow fear and hesitation to slow our progress while other industries race ahead. For the sake of our members, the people whose financial futures depend on our decisions, we must choose progress.
The AI revolution isn’t coming to the retirement industry. It’s already here. The question is: are we ready to harness its power responsibly?
