At trade publication Commercial Observer's Power Tech & AI Forum held at Marx Realty's 10 Grand Central building, recent developments in artificial intelligence took centre stage for the real estate industry. Technology and real estate professionals discussed how automated systems are already accelerating transaction speeds and which technological leaps lie ahead in the near future.
Tom Acitelli, deputy editor of the Commercial Observer, opened the event by noting that numerous innovations over the years had promised substantial disruption. While concepts such as the metaverse failed to meet expectations, artificial intelligence is triggering a profound transformation across the sector unmatched by any prior trend.
Efficiency gains in document review and pre-construction
During the panel discussion on accelerating transactions, Abhinav Somani, founder of Cadastral and head of real estate at Legora, pointed to the technology's primary current benefit: the structured processing of extensive document sets. Legal departments and law firms are deploying specialised legal AI tools to expedite negotiations and contract reviews. Joseph Sipala, managing director at JLL, confirmed these efficiency gains. By integrating systems such as Legora and Claude, the company requires only a quarter of the original time for portfolio analyses and reviews of individual leases, reducing a working week to a single day.
Zach Aarons, co-founder and general partner of venture capital firm MetaProp, highlighted advances in output accuracy. His fund is investing substantial resources in applications for the pre-construction phase. Combining computer vision algorithms with generative models allows millions of pages of construction documentation and bills of quantities to be cross-referenced with precision. Portfolio company Handoff determines material quantities for residential projects with an accuracy of 97 per cent, compared with 94 per cent for human teams, with assessments carried out twenty to fifty times faster.
Specialised models and physical automation
The technology is also opening up new opportunities for developers. Nick Durst, principal and director of operations at The Durst Organization, explained that data rooms in land acquisitions can be analysed systematically within minutes to understand properties in detail.
Looking ahead to the next one to two years, the panellists expect an increasing differentiation of tools:
- —Replacement of off-the-shelf applications with tailored systems that incorporate institutional corporate knowledge.
- —Combination of different models to cover specific workflows in a cost-optimised manner.
- —Market readiness of humanoid robots from 2027 featuring 3D computer vision, speech and sensory capabilities.
Somani emphasised that the transition from text-based systems to physical robotic units operating in three dimensions will have more far-reaching consequences for the real estate sector than the mere automation of administrative tasks.
