AI, MIT and the next real estate cycle

Mr. Cassano is Partner, National Real Estate Leader, for PwC Canada.

Between June and September 2026, my colleagues and I sat down with investors, owners, developers, lenders and operators across the country for the PwC and Urban Land Institute (ULI) 2027 Emerging Trends in Real Estate® (ETRE) report, now in its 22nd year in Canada. Across more than 250 interviews, one message came through clearly: The near-term priority is operating excellence, using AI and better processes to run today’s assets better.

Midway through that season, I spent an afternoon at MIT with Steve Weikal, who chairs the university’s Real Estate Transformation Lab and is general partner of MET Fund II, which invests in early-stage MIT-connected companies. If the interviews showed me where the industry is focused today, that afternoon showed me what comes next: AI changing not just how real estate is run, but what real estate is, as an asset, an industry and an investment class.

Consistent with PwC’s Value in Motion research, Steve treats materials, construction and housing delivery as one connected system rather than separate disciplines. That lens could shape how the real estate industry evolves over the next decade as it absorbs megatrends like climate, demographic shifts and technology change.

Three ideas came into focus:

1. Reimagining the asset itself

Steve sees three levels of AI adoption in real estate:

• Level 1 — Automate: Use AI to do the same task faster and cheaper. Automated lease abstraction, faster valuation models, quicker due diligence.
• Level 2 — Augment: Use AI to help humans make better decisions. Predictive maintenance, revenue optimization, tenant analytics, portfolio scoring.
• Level 3 — Reimagine: Use AI to change what the asset is and how it earns. New asset classes, new revenue streams, new ways to build and monetize buildings that were not viable before.

Across this summer’s ETRE interviews, the message is remarkably consistent: the near-term focus is operational efficiency and excellence. That means automating the back office, tightening asset-level performance, and using AI to squeeze cost and time out of workflows that have been fundamentally the same for decades (levels 1 and 2 in Steve’s framework above). Our interviewees have shared many examples of measurable productivity gains, cost takeout, quality improvements and better experiences for employees and tenants.

MET Fund II backs companies operating at Level 3—ventures reimagining what the asset is from the ground up. Steve frames it as the “Built Environment Transition”: a generational shift in how we finance, design, build, operate and monetize the physical world.

The Lab’s research shows what Level 3 looks like in practice. Materials science is producing construction inputs that generate energy or sequester carbon. Buildings are being designed as grid-connected systems, not just enclosed spaces. And operating platforms are turning buildings into revenue sources beyond rent—for instance, energy sales. All three assume AI is the enabling engine.

Physical AI is where these levels start to connect. In September 2026, PwC Canada launched a Physical AI offering focused on technology that lets robots and autonomous systems sense, reason and act in the real world, and the read-across to real estate is direct.

At Level 2, drones and robots inspect roofs and envelopes, autonomous systems run building operations, and digital twins test those systems before they go live, so today’s assets run safer, leaner and better instrumented. At Level 3, the same technology moves onto the factory floor, where robotic production lines build the asset itself.

As Subo Chatterjee, PwC Canada’s Physical AI and emerging technology leader, puts it, Physical AI is “a generational opportunity to drive a step change in Canada’s productivity equation.”

The Level 3 shift is early in North America, and it is uneven. But at scale, it already exists. In Japan, Sekisui House produces tens of thousands of factory-built homes per year on industrialized production lines—homes as configurable products rather than bespoke projects, with integrated energy systems and ongoing revenue streams, including rooftop battery storage and long-term maintenance contracts, layered on.

The strategic question for our industry is how to run both playbooks at once: capture the near-term productivity gains while positioning for the reinvention.

At the frontier, real estate-adjacent examples show how new asset classes get created in the first place. GaiaAI, an MIT-Harvard startup, uses LiDAR, computer vision and satellite data—the same AI stack that powers autonomous vehicles—to turn timberland into a more investable, data-rich real asset. It is a new asset class made possible by AI, and it points to how AI reimagines the physical world.

2. Materials science and the convergence of real assets

ETRE has been tracking the convergence of real estate and infrastructure for a few years now. When a building’s envelope generates its own power, its structural material stores energy and its systems participate in the grid, it stops being just real estate—it becomes a real asset that is part building, part infrastructure, part technology platform.

As Steve put it, software layers commoditize while physics compounds.

What’s new is how investable this convergence is becoming. Two developments stood out from our conversation:

• The building envelope is becoming the technology stack. Active Surfaces, an MIT-connected venture, builds thin-film solar laminates with efficiency comparable to silicon solar panels, while being substantially lighter and faster to install. Active Surfaces reports that over 60 per cent of warehouse roofs cannot support conventional glass solar arrays—a constraint its thin-film format is designed to solve, turning the building envelope itself into a power source.

• Capital is following returns toward these hybrid real assets. Steve sees capital deploying faster than the supply of assets. Real assets money wants long-duration, cash-flowing, carbon-conscious assets, and it is comfortable underwriting technology risk alongside physical risk. PwC’s US Deals 2026 midyear outlook confirms the shift: investors are increasingly underwriting assets “as integrated operating and infrastructure platforms rather than standalone property investments.”

3. From bespoke to scalable: housing as a platform

Steve argues that multifamily has been built the way cars were built before the assembly line—one at a time, on site, bespoke. Months of design revisions, schedules exposed to weather and material volatility, and know-how that rarely carries to the next project. Could AI change that?

MIT is exploring what development looks like at scale. A developer selects from a product catalogue of pre-engineered building platforms—the way an automaker offers platforms that scale across models, except now the building is the product. AI configures to site, unit mix and code compliance in weeks. Components are manufactured indoors to millimetre tolerances. On site, assembly is measured in weeks rather than years.

This vision is already happening. MIT-founded Reframe Systems operates robotic micro factories. Juno Residential has productized mass-timber multifamily around factory-fabricated components, first delivered in Austin, Tex. Interior fit-out is being productized, too: Kit Switch, a Stanford University-founded startup, delivers plug-and-play kitchen and bathroom modules that install in a day instead of weeks.

AI is the capability that removes the bottlenecks: design-to-manufacturing translation, site-to-site variation and supply chain optimization.

The Canadian opportunity

I think investors should be looking at Canada as one of the more investable markets for this shift. Tailwinds worth noting:

• Government alignment: What I’m seeing on the ground is genuine co-ordination across federal, provincial and municipal housing initiatives—more concrete than it has been in years.

• Natural advantages: Canada has natural advantages in the inputs this transition depends on—mass timber, critical minerals and low-carbon energy.

• Infrastructure at generational scale: PwC Canada forecasts US$4.7 trillion in cumulative infrastructure spending in Canada between 2024 and 2050 and makes the case for moving from isolated assets to integrated, multi-use systems. That is the convergence Steve describes, at national scale.

• Capital already committing: The first Canada Investment Summit, held in Toronto in September, generated nearly $500 billion in new investment commitments, including the $50 billion CPP Investments–Brookfield Maple Fund for critical infrastructure. Add Canada’s combined housing and defence commitments of more than $100 billion over five years, much of it targeting exactly what this transition requires: industrialized construction, dual-use infrastructure and domestic supply chain capacity.

I also note that global industrialized builders are noticing Canada. SEKISUI Chemical, maker of factory-built Sekisui Heim homes in Japan since 1970, has opened SEKISUI Canada Modular Solutions in Burlington, Ont., a factory producing steel-frame modules for residential and commercial projects. Another of Japan’s factory-built housing pioneers now has a production line in Ontario. What’s encouraging is the choice: one of Japan’s most experienced factory builders has made Burlington its Canadian base.

To me, that is a signal of confidence in Canadian housing demand and in the policy direction behind industrialized construction.

What MIT is seeing and what Canada is building are converging. ETRE 2027 is putting the operator at the centre—and rightly so for this market. Steve’s lens is what comes after. The operators who win the next cycle will be the ones who reinvent the asset itself—through better materials, industrialized construction and AI-driven design and delivery, while governing the risk that comes with it.

Fred Cassano is Partner, National Real Estate Leader, for PwC Canada. Fred’s clients include public and private real estate landlords, developers, construction and engineering companies, hotel managers and owners, retailers, and high-net-worth individuals and family offices. He is on the Canadian editorial team for Emerging Trends in Real Estate, a thought leadership report that PwC publishes annually with Urban Land Institute. Fred is a Chartered Accountant and a graduate of York University’s Schulich School of Business.

The Canadian Family Offices newsletter comes out on Sundays and Wednesdays. If you are interested in stories about Canadian enterprising families, family offices and the professionals who work with them, but like your content aggregated, you can sign up for our free newsletter here.

Please visit here to see information about our standards of journalistic excellence.