How property sector can embrace collaborative AI

The rise of artificial intelligence (AI) has sparked much excitement in the property sector, presenting us with a host of opportunities. 

Yet as we hit our stride in an AI-driven world, it's important that we recognise a major strength of this technology – namely, that it is by nature distributed and collaborative. 

This not only shapes the way we develop AI, but also how we engage in dialogue about its benefits and risks, fostering opportunities for collaboration and prioritising coordination over control in property sector frameworks.

Even established technology players have been wrongfooted by the rapid proliferation of AI. For instance, Google's urgent mobilisation of its AI model Gemini met a swift backlash, followed by a fall in share prices that illustrates the commercial risks of its haste. 

The property sector should therefore recognise that, while AI offers benefits and possibilities, a sales-heavy, product-centric approach by vendors creates risks at the interface between property and technology. 

Indeed, even nimble tech start-ups must still walk the line between transparency and credibility on the one hand and the protection of the valuable IP on the other.

Property advisory firms employing statistical models and methods to uncover opportunities and mitigate risks must ensure these analytical tools can be explained; simply acquiring and deploying the latest AI tools is not enough.

So as AI-powered systems and methodologies gain traction, it is a combination of their learning capabilities with the open exchange of expertise across trusted networks and partnerships that will enable lasting competitive advantage.

The prevalent model of software as a service (SaaS) used in deploying AI presents its own set of challenges. While SaaS offers ease of access and lowers upfront costs, it often locks users into specific platforms and data usage policies.

Indeed, SaaS models may inhibit innovation because they typically operate on closed systems, where proprietary algorithms dictate functionality and limit interoperability with other platforms.

This discourages the bespoke adaptation that could otherwise enable forward-thinking approaches tailored to specific property sector needs. Then there is a more human dimension – excessively handrailed user journeys will limit its potential by failing to stimulate their imagination.

AI instead deserves adaptable tooling, as well as the skilled practitioners capable of operating it. We certainly need to invest in SaaS; but we also need to develop people and capabilities that allow an array of open-source AI methods to be synthesised.

In addition, we need to attract and retain more people in property who can develop a bespoke technical solution to a client's particular problem.

Think of a valuation model that specialists can configure to incorporate sentiment data from social media, or a retail site selection model that is improved with dynamic traffic data.

Our ambitions cannot ignore data quality considerations either – AI will not close process gaps or address behavioural shortfalls that deprive us of useful insights about markets, assets, transactions and customers.

Just as AI development takes place when these diverse parties, datasets and algorithms come together in an iterative fashion, distributed networks similarly depend on the aggregation of interconnected nodes that exchange information, resources or services.

This allows for greater adaptability, resilience and scalability compared to more rigid, hierarchical structures. AI will therefore thrive on collective effort, drawing from a vast pool of data, expertise and open-source resources such as the platform Github.

Within a distributed and multi-nodal framework, property businesses can more readily experiment with and iteratively refine AI tools, aligning more closely with their particular expertise.

Diverse use cases for AI have already been showcased across the property sector, with applications including predictive building maintenance, automated tenant screening, devising dynamic pricing strategies, and reconciling large arrays of viability and sustainability data.

Successful approaches commonly draw on process automation and rapid scalability, and these efforts rarely emerge from siloed, proprietary resources. Siemens, for instance, drew on open-source libraries and frameworks during the development of its Building X platform, which optimises HVAC usage to cut costs for occupiers.

Indeed, the most persuasive technical case studies involve real-estate practitioners using technology to provide differentiated but highly personalised services.

For example, in solving the problem of fertiliser run-off from a single farm, Knight Frank worked with the Environment Agency to open up a new data-driven market for nutrient offsets, enabling housing development across the wider river basin.

Bespoke approaches such as these can be scaled nationally, allowing a multitude of partners to scan for similar opportunities.

Those who can artfully blend technology with collaborative human approaches will ride the next wave of innovation and prosperity.

As we navigate this frontier, RICS – as our regulator – has a key role to play in enabling and expanding dialogue around AI in collaborative forums such as the Tech Partner Programme. 

AI is a tool we must engage with thoughtfully – but we needn't jeopardise our intellectual property to discuss ideas and talk about our common goals. Shared concerns include attracting the right talent from universities, how to deploy new skills across the sector, ensuring data supply and quality, and maintaining transparency and public confidence.

RICS must also foster a critical and dispassionate approach so that its membership can use AI objectively. The organisation can also ensure that, as we make greater use of AI, we do so with a keen awareness of our professional responsibilities and the broader societal implications.

AI's distributed nature encourages us to break down silos, fostering cross-disciplinary partnerships that draw on diverse expertise. This collaborative approach is not merely beneficial but essential, as the complexity of real-estate challenges demands a multidisciplinary perspective.

As the property sector continues to evolve, market players must champion collaborative efforts and open approaches to harness fully the potential of distributed AI systems.