Something shifted in housing this year. AI stopped being a conference topic and started turning up in the actual work. The problem is that adoption has raced ahead of understanding, and a lot of teams are now running tools they can't quite explain.

The numbers make the gap hard to ignore. The 2026 NHF/Phoenix survey found 47% of housing associations already using AI in day-to-day operations, with another 23% planning to. Yet 87% reported low knowledge of AI, and 44% had no AI policy at all. So the sector is using a technology it mostly doesn't understand, without the rules to govern it. That is a recipe for quiet mistakes: a triage decision no one can account for, a resident flagged or missed for reasons no one can trace.

This is why the conversation is turning from "are you using AI" to "can you stand behind what it does". By the end of 2026, AI in social housing will be judged less by how widely it is used and more by how confidently it is governed. For a repairs team, that means the useful question is not how clever the model is. It is whether you can see why a job was prioritised, what information the decision was based on, and whether a person could check it. Trust comes from transparency, not novelty.

Good AI in this setting is boring in the best way. It does one clear job, shows its working, and keeps a record. When a repair is reported, the resident's own words, the priority, any vulnerability, all sit in one place where a manager can review them. Nothing happens in a black box. If the Regulator or the Ombudsman asks how a decision was made, the answer is in the file, not in someone's memory.

That is the approach behind Alix. Alix structures each repair at the point it is reported, flags what is urgent and who is vulnerable, and holds the full picture, notes and photos included, in a single system. Every step is visible and reviewable. It is not AI you have to take on faith. It is AI a team can supervise, which is exactly what a sector with low confidence and thin policies actually needs right now.

The associations that pull ahead will not be the ones that adopted AI fastest. They will be the ones who can explain it. If half the sector is already using these tools, the advantage now goes to whoever can govern them well, and show their working when it counts.