Artificial intelligence is already embedded in many areas of commercial real estate, from drafting reports to processing documents and supporting marketing campaigns. Much of that activity relies on generative AI tools that respond to human prompts and generate new content from large data sets.
A recent McKinsey analysis argues that the next phase for the industry is agentic AI, which shifts the focus from content creation to decision-making and task execution. According to the report, agentic AI can be embedded directly into core real estate operating systems and deliver support across four key domains: maintenance, leasing and renewals, investments and asset management, and construction and capital expenditure planning.
Generative and agentic AI differ in how they operate. Generative AI systems draw on machine learning models to identify patterns in data, then produce text or other outputs when users give specific prompts. By contrast, agentic AI systems are designed to pursue defined goals with limited supervision, making decisions and taking actions without requiring constant human input.
IBM characterizes generative AI as reactive, because it needs prompts to generate content, while agentic AI is described as proactive. Under that framework, agentic AI can accept a task, operate against that objective and, in many cases, continue executing without ongoing human oversight, as long as appropriate guardrails and governance are in place.
Within maintenance operations, McKinsey suggests that agentic AI can monitor systems, flag issues such as leaks before they escalate, coordinate vendor responses and keep tenants informed. This combination of early detection, automated coordination and communication can reduce response times and limit physical damage.
On the leasing side, agentic AI can take on workflows like scheduling tours, processing applications and managing renewals. Automating these repetitive steps can free on-site and back-office teams to concentrate on tenant relationships and retention efforts, with the goal of improving renewal performance and speeding up lease-up.
For investments and asset management, the report notes that agentic AI can gather data from multiple sources, assemble reports and surface information needed for portfolio decisions. With routine data work handled by AI, investment and asset management teams can devote more time to higher-level strategy and portfolio optimization.
In construction and capital projects, agentic AI can help manage documentation, interpret compliance requirements and coordinate project workflows. By structuring information and tracking progress, the technology can support earlier identification of risks and help keep schedules and budgets on track.
McKinsey analysts indicate that, over time, agentic AI systems can learn from each cycle of work, improving performance and gradually optimizing building and portfolio operations. Potential benefits highlighted include better tenant experiences, more efficient capital project delivery and more informed asset management.
The report also stresses that realizing these benefits requires time, investment and what it calls a disciplined architecture. That includes clear data ownership, feedback and learning loops, and careful design of how AI connects to existing systems. Finally, the authors emphasize that agentic AI should be used to assist and augment human expertise rather than replace it, reinforcing the need for human oversight in setting objectives and reviewing outcomes.
