Source: Unite.AI
Heimstaden is using ChatGPT Enterprise and Codex to rework home lettings, environmental-risk analysis and property operations, OpenAI said on November 3, 2026, reporting a reduction of more than 99% in Dutch property-image production costs.
The European residential real estate operator accelerated its AI program in June 2025 and selected OpenAI after evaluating several platforms. Its strategy combines employee training and tool access, a central transformation office supported by a network of AI champions, and preparation of property-management systems and data models for AI. The aim is to redesign complete business processes while employees retain responsibility for decisions.
AI Across the Letting Workflow
When a home becomes vacant, Heimstaden’s letting work spans rental valuation, photography, advertisements and responses to market demand. AI now supports pricing recommendations through tools built in ChatGPT Sites, styles property images and writes apartment descriptions. An AI agent reviews live advertisements against the company’s quality standards, while AI-generated market feedback helps letting teams take corrective action and make pricing more dynamic.
Staff remain the decision-makers, with AI handling more of the analysis and production work leading up to their approvals. That changes the workflow surrounding a vacancy without removing employee review from the process.
Heimstaden said the cost of producing a virtually staged property image in the Netherlands fell from approximately €50 to less than €0.10, and estimated annual savings potential of €130,000 in that market alone. The saving sits within a broader letting redesign that coordinates pricing, advertising, listing review and market feedback rather than treating them as separate tasks.
Codex Supports Environmental-Risk Analysis
Beyond lettings, environmental sustainability analyst Julia Erlandsen used Codex for a portfolio-wide environmental-risk assessment despite not having a software-development background. The necessary information was spread across public sources, formats and datasets.
Erlandsen used Codex to find, clean and standardise the data, check its quality and create processing pipelines that could be run repeatedly. She retained control of the methodology and validated the work herself, keeping the analytical decisions with the domain specialist.
OpenAI reported that the project assessed 20,112 assets and generated more than six million rows of useful data. The estimated delivery requirement fell from five people working for a year to two people working for two months, and Heimstaden avoided engaging external consultants. The project is becoming a repeatable capability for climate-risk assessments, reporting requirements and investor decision-making, with Heimstaden owning the methodology and able to develop it further.
Employee Controls Shape the Next Systems
ChatGPT Enterprise launched at Heimstaden in September 2025. OpenAI reported that half of employees were using it within two to three months. The central transformation office maintains a library of use cases so staff can learn from examples elsewhere in the business. Employees use approved tools within data-security rules and must review AI-generated content before sending it.
Adoption also extends beyond office-based analysis. In Norway, a caretaker used ChatGPT to assess a neglected apartment and turn observations into a prioritised refurbishment plan, which the caretaker could review and refine. The example illustrates Heimstaden’s intended division of labour: AI organises information and proposes a plan, while the employee remains responsible for the outcome.
Heimstaden’s roadmap moves that model into core business systems. Its Procurement Intelligence Platform, currently being rolled out, combines workflow automation with OpenAI models accessed through APIs. Designed to operate across Outlook, SharePoint and Teams, it supports contract analysis, negotiation preparation and contract lifecycle management. Employees retain responsibility for commercial decisions and approvals.
Heimstaden is also co-developing Amplio Estate, a property-management system intended to support AI-powered workflows through MCP. Its goal is to let employees use AI such as ChatGPT to retrieve property information and take action in core business systems with human oversight. This work supports the company’s longer-term ambition for increasingly self-operating buildings, where AI helps anticipate needs, coordinate work and support decisions across the property lifecycle.
The roadmap extends AI from producing listings and analysis to taking action inside operational systems. That makes employee oversight central to Heimstaden’s plans: its vision of increasingly self-operating buildings still assigns review, validation and approval to the people responsible for homes and customers.
