Artificial intelligence is rapidly gaining traction across the commercial real estate (CRE) sector, but its full potential will depend heavily on how the industry manages and shares data, according to new analysis by Colliers.
The property consultancy has outlined six broad categories of AI, noting that while the technology spectrum is wide, most current CRE applications are concentrated in only a few areas. The firm’s latest paper focuses on the segments already in use or expected to see near-term adoption, while also examining how the full range of AI capabilities could be deployed in future.
At the centre of the discussion is transparency. Data across CRE companies remains highly fragmented and often treated as proprietary, creating a structural barrier to wider AI adoption. Analysts say meaningful progress will require a cultural shift towards greater standardisation and controlled data sharing.
The quality and breadth of data are also critical. Limited or biased datasets can distort AI outputs, making comprehensive, cross-organisational data essential for reliable benchmarking and forecasting. Without this foundation, the risk of inconsistent or misleading insights increases.
Governance frameworks are emerging as another priority. Industry participants are being encouraged to develop formal data governance programmes and structured content libraries. These systems are expected to reduce risk, improve accuracy and consistency, and enable more effective use of AI tools.
The CRE sector is not short of raw information. It generates vast volumes of data spanning markets, leases and transactions — all valuable inputs for training advanced AI models. However, experts warn that these datasets must be carefully curated to avoid bias or gaps that could undermine results.
Effective data management will ultimately determine success. This includes maintaining data integrity, regulating access permissions and establishing robust data-sharing agreements among stakeholders. As AI adoption accelerates, the industry’s ability to organise and govern its data ecosystem may prove just as important as the technology itself.
