Housing associations face cost pressure, heavy repairs loads, and strict consumer regulation — a context where AI can help, if tenant fairness is protected. Here’s how in 2026. (dgm implements osFoundry as an independent partner; service responsibility stays with the provider.)
Where AI helps
- repairs triage and predictive/planned maintenance (analysing sensor + historical data);
- damp-and-mould risk identification;
- tenant enquiry handling;
- complaints management; and
- back-office automation.
Repairs and predictive maintenance is often the strongest use case given the scale of repairs costs.
The regulation
Housing associations are registered providers regulated by the Regulator of Social Housing (RSH). The frameworks that shape AI:
- the new consumer standards (in force since 1 April 2024);
- Tenant Satisfaction Measures (TSM); and
- the requirement for fairness, transparency and access to human decision-making.
UK GDPR applies to tenant data.
Tenant-facing AI: fairness and a human route
This is the distinctly UK constraint. RSH consumer standards demand fairness, transparency, and a route to human decision-making. So tenant-facing AI must not produce unfair outcomes, must be transparent, and must let tenants reach a human — making human oversight and clear escalation essential, not optional. (See human in the loop.)
Where osFoundry and dgm fit
dgm builds fairness- and transparency-aware AI on osFoundry: human escalation built into tenant-facing workflows, audit logging for transparency, and data control (self-hosting or an EU region — it publishes US/EU/JP regions, not a UK one) for sensitive tenant data. Repairs/predictive-maintenance use cases pair well with its retrieval and workflow capabilities.
dgm is an independent integration partner with zero integrations so far — no provider case studies to claim. Service responsibility stays with the provider. To scope a standards-aware housing AI project, book a consultation with dgm. Not regulatory advice.