AI Impact on Offices

How Might the Office Shift from Work Container to Decision Hub? 

THE SHIFT

1. The value of bringing people together to make decisions, collaborate, and align priorities;
2. The ability of organizations and managers to create oversight and consistency;
3. The office’s role as a base for customer interactions and a manifestation of brand

THE PREDICTION

REAL ESTATE IMPLICATIONS

The more important question is not simply how much space is needed in any given market, but what offices are optimized to do. As AI reduces the effort required to produce work, the role of the office becomes less about supporting individual task completion and more about accelerating decisions that require human context and shared commitment. The premium moves toward settings that increase decision quality and speed: smaller groups, clearer facilitation, better capture of conversation, and faster conversion from discussion into agreed actions. In that context, meeting rooms become production environments - designed to generate outputs, not just host dialogue.

For occupiers, the question moves from ‘how many desks do we need’ to ‘which decisions do we need to make faster and better’ - and whether the office is designed to consistently produce those outcomes. For investors, the question becomes which assets are positioned to support that shift - and which are still priced on a model of work that AI increasingly compresses. 

What Will an AI-Ready Office Look Like?

THE SHIFT

Since the earliest modern offices, fit-out has been designed around people and furniture: desks, meeting rooms, and amenities that support presence.  Technology has dictated many of these requirements: typewriters and paper needed filing cabinets, telephones promoted cubicles, whereas the cell phone untethered workers and today’s peripherals and monitors shaped the standard desk.  This has had a central role in reshaping the perception and layout of the office over time. 

As AI becomes embedded in everyday work, the form of the office will again shift.  It will start to behave less like a static container and more like an integrated element of the work process - one that supports AI-led workflows and makes work easier to capture, retrieve, and reuse.

THE PREDICTION

Many of the practical benefits of AI depend on inputs: what was said, what was decided, what was shown, and how teams progressed from discussion to action. That shifts the functional brief for office space. A larger share of value moves into settings designed for high-quality interaction and output: small decision rooms, project ‘war rooms’, and studio-like spaces that support recording, playback, and remote participation. The objective is not surveillance. It is to turn live work into a usable record: searchable conversations, clear decisions, defined actions, and outputs that can be revisited and built on - in turn compounding organizational knowledge.

REAL ESTATE IMPLICATIONS

This also changes what sits on the floorplate. If more interaction is mediated through AI (including digital assistants and, over time, more visual interfaces such as avatars), the workspace needs different peripherals and infrastructure. Acoustics become more important. Lighting becomes more intentional. Screens, cameras, and collaboration surfaces become core workplace utilities. Reliable connectivity becomes less a convenience and more a prerequisite. Now with expanded use cases, sensors and data capture start to play a role in optimizing how space performs: room utilization, environmental quality, and the conditions that support focus and collaboration. In effect, the next-generation office is designed less like a floorplate and more like a system - optimized for inputs, outputs, and performance.

The deeper implication is that ‘AI readiness’ is not only a software decision. It is a fit-out and infrastructure decision. For occupiers, the question becomes how to design space that consistently produces high-quality interactions and reusable outputs. For investors, it becomes which buildings can support this layer of digital infrastructure - and which will require expensive reinvention to remain competitive. This creates a pathway to differentiation: assets that can credibly support capture-quality collaboration and digital infrastructure should command stronger demand and retention

Will AI Drive New Office Bifurcation?

THE SHIFT

Not all office work is equally exposed to AI. Over the next decade, AI is likely to compress some formats of white-collar activity more than others - particularly office ‘factory’ functions built around standardized tasks at scale, such as shared service centers, large back-office processing, and certain call-center type operations. In those environments, the economic case for AI is more straightforward because demand for the output tends to be more inelastic - you don’t usually need more back-office processing just because it’s cheaper.

That exposure has two consequences. The first is at the level of product. Building formats designed primarily to host high-volume, process-driven work are likely to be less resilient if the underlying activity becomes less space-intensive. The second is at the level of location. Some cities and submarkets have become specialized hubs for these functions, often because of labor arbitrage, cost advantages, and large pools of operational talent. If AI reduces the need to place routine execution work in those locations, it can become a structural demand headwind.

THE PREDICTION

And so bifurcation becomes both a market story as well as a product story. Cities with demand anchored in higher-value decision-making, complex client work, and the industries that rely on dense networks should remain more resilient, even as roles and workflows change. Markets that are disproportionately concentrated in routine execution functions may face a sharper adjustment - not because the offices themselves are deficient, but because the specific economic purpose those offices served becomes easier to deliver with fewer people and less space.

REAL ESTATE IMPLICATIONS

As with previous industrial shifts, the strategic question for exposed markets is reinvention. If the value of routine processing declines, what replaces it? Some locations will move up the value chain into higher-skill roles and specialized services. Others will diversify into adjacent sectors, build new institutional strengths, or reposition toward industries that benefit from local presence. The point is that AI will not hit all office markets evenly. It will amplify differences in the underlying work each market is selling to the world. Cities that treat this as an upgrade moment (skills, sectors, and institutions) can turn exposure into a repositioning opportunity anchored for instance in the growth of AI-adjacent industries. 

For occupiers, this raises a network question: which functions belong where in an AI-enabled operating model. For investors, it raises a market selection question: which locations are priced for stable demand, but exposed to structural substitution?

Will Flex Become Central to Portfolio Strategy?

THE SHIFT

Flex has often been treated as a cyclical response to uncertainty - a short-term bridge when demand is unclear or headcount is in flux. That will remain true in the near term, while the impact of AI on an organization’s operating model and occupational needs remains unclear and untested. But over the next decade, AI is also likely to make uncertainty more persistent in a different way. Enduringly faster decision cycles, faster iteration, and faster organizational change increase the value of optionality.

THE PREDICTION

As AI compresses execution and reduces the time required to test ideas, organizations can scale projects up and down more quickly. Teams form and disperse faster. New initiatives can be launched with fewer barriers. That creates a different spatial requirement: not simply ‘less’ or ‘more’ office, but greater variability in how space is provisioned over time.

REAL ESTATE IMPLICATIONS

This is where flex moves from a short-term occupier hedge against uncertainty, to a structural component of portfolio design. For many organizations, a blended model will become more attractive: a smaller core footprint optimized for decision-making and culture, supported by flex capacity that can absorb volatility - growth spurts, project teams, market entries, and reorganizations - without locking in long-term commitments.

There is also a market implication. Flex typically carries a higher unit cost than conventional leasing. Some organizations will absorb that premium because of the value it creates in optionality and speed. Others will offset it by redesigning the portfolio: reducing fixed commitments, tightening the core footprint, and improving utilization over time. In effect, flex changes the trade-off: higher price per unit of space, in exchange for lower risk and greater adaptability.

The implication is also material for landlords and investors. Flex is not just a leasing product; it is an operating model. Performance depends on service quality, experience consistency, and the ability to manage churn - capabilities that increasingly resemble hospitality as much as traditional office asset management. In some markets, owners will partner with operators. In others, they will build capability in-house. Either way, flex becomes a source of differentiation; it adds a portfolio tool that improves resilience.

For occupiers, the advantage is speed and risk management: access to the right space when it is needed, without overcommitting. For investors, the question is how flex changes income profiles, operating requirements, and building design; and which assets are positioned to support that model

Does AI Create New Market-Fit Risks?

THE SHIFT

For much of the last cycle, office product has been designed for an ‘average tenant’: a broadly similar set of needs around desks, meeting rooms, and amenities, delivered through a familiar specification. Even where buildings were differentiated by quality, many assumed the demand base would remain reasonably consistent, with most occupiers wanting broadly the same thing, just in different grades.

THE PREDICTION

AI increases heterogeneity. As organizations adopt AI at different speeds, and as AI changes workflows in different ways, demand becomes more uneven across industries, functions, and risk postures. Some firms will double-down on in-person collaboration and client experience. Others will redesign work to be more distributed. Some will require high levels of capture and secure digital infrastructure. Others will prioritize cost and flexibility. The result is not a single ‘new office’, but a wider spread of requirements.

REAL ESTATE IMPLICATIONS

That creates market fit risk for office product. Buildings designed too narrowly around one assumed tenant profile become harder to lease, because the pool of perfect matches shrinks. At the same time, generic ‘one-size-fits-all’ product can struggle if it does not excel at the few things that matter most. The winners are assets that can support multiple use-cases without constant reinvention: adaptable floorplates, credible infrastructure, and the ability to offer different settings and service levels as tenant needs evolve.

It also changes how development and landlord works are approached. As requirements become more varied, the value of pre-letting and occupier-led schemes can rise - not only to secure income, but also to reduce specification risk. Where that is not possible, phasing becomes a risk management tool: deferring elements of fit-out until demand is clearer, and keeping more of the building ‘option-ready’ rather than over-committed to a single assumed layout. This is particularly relevant for items that are expensive to change later, such as core MEP capacity and distribution, risers and comms pathways, acoustic performance, ceiling and lighting strategies, and the allocation of space between collaboration settings and conventional work points.

For occupiers, this shift increases the value of choice and optionality, but also makes portfolio decisions more forensic. For investors and developers, it sharpens the underwriting question: not only ‘is this a good building’, but ‘how broad is the tenant market it can reliably serve’ - and how much capital is being placed at risk before that demand is visible.