Trust is becoming one of the most important currencies in South Africa’s property ecosystem as artificial intelligence (AI) moves from analysing information to making and executing decisions.
That was the central theme which emerged from Lightstone’s recent 20 Years in Banking – The Trust Algorithm conference in Johannesburg.
For the past two decades, data and automation have steadily changed the way South Africa’s banks, insurers and property businesses make decisions – and the next 20 years could be far more disruptive.
AI has moved from helping analyse information to recommending decisions, initiating processes and potentially acting itself. This means that data accuracy is being surpassed by other questions, such as: can I understand why this decision was made? Who authorised it? Can I challenge it? And can I reverse it?
Technology analyst Arthur Goldstuck said that as AI moves into its agentic phase, accountability becomes critical. Every automated action needs to be traceable, explainable and reversible. This has profound implications for property and financial services, where technology increasingly influences decisions about valuations, lending, risk and, ultimately, people’s homes.
More data doesn’t automatically mean better decisions
The irony is that organisations have access to more information than ever through increasingly sophisticated models, APIs, automation and AI. Yet, as Anthony Miller, co-founder of Lightstone and CEO of Simply, put it: more data doesn’t necessarily mean more trust. The real differentiator is the quality of the intelligence created from that data.
For Lightstone, that means accuracy must be non-negotiable, methodologies robust, and decisions auditable and explainable. Technology is only part of the equation.
“We have learned over 20 years of working with banks, insurers and property businesses that trust is built through behaviour over time. It is created when partners deliver consistently, challenge one another when necessary and are prepared to share information because they have confidence in how that information will be used. That becomes even more important as AI increases the speed and scale at which decisions can be made,” says Sara Winstone, Managing Director of Lightstone Property.
The opportunity is bigger than AI
One of the most exciting possibilities for the next era of property intelligence is to move beyond building better algorithms to connecting an ecosystem that is still fragmented.
Municipalities, valuers, banks, insurers, Estate Agents and consumers all hold different pieces of the property picture. Connecting those signals could enable faster, more informed and potentially fairer decisions.
Around 60% of households sit outside the Deeds Registry and so remain largely invisible to traditional South African property market systems. New sources of intelligence, including aerial imagery, computer vision, municipal information and alternative credit data, create the possibility of understanding those properties and communities far better.
“That matters because better visibility is not simply a data problem but could help create a path towards formal funding for people whose homes and assets have historically been difficult for the financial system to see,” says Winstone.
From reacting to anticipating
The same principle could fundamentally change the way lenders manage risk. Today, financial distress is often identified once a customer has already missed a bond repayment. A more connected intelligence ecosystem could identify early signals, such as unpaid municipal rates or changes in financial behaviour, allowing institutions to engage with customers sooner and with greater context. The goal should not be more automated decisions for their own sake but better decisions, made earlier, with greater understanding of the person affected by them.
RMB Chief Economist and Head of Research Isaah Mhlanga made a related point: in a rapidly changing economic environment, traditional indicators can lag reality. More timely data and AI-enabled analysis can help organisations understand what is happening sooner, but only if the underlying information remains reliable and relevant.
Human judgement becomes more important, not less
Perhaps counter-intuitively, greater automation may make the human element more valuable. Nedbank’s Josephine Baliah argued that AI will not replace human ingenuity and that trust continues to come from human connection, aligned values and how organisations respond when things go wrong.
That is an important distinction. AI can analyse more information than any individual, identify patterns and automate processes at enormous scale, but people still provide judgement, accountability, empathy and context. The future isn’t about choosing between human expertise and technology, but about designing systems in which each makes the other better.
The next 20 years
For Lightstone, the opportunity is to move beyond individual valuations, reports and data feeds towards a trusted intelligence layer connecting the property ecosystem. That could mean decisions are continuously monitored rather than made once; a more complete view of South Africa’s formal and informal property market; earlier identification of financial distress; and increasingly automated decisions that can still be understood and interrogated.
While the technology to enable much of it is arriving quickly, the harder challenge is to earn the trust to use it because the organisations that shape the next 20 years of property will not necessarily be those with the most data or even the most sophisticated AI. They will be the ones people trust to turn that intelligence into better decisions.
