On 9 July 2026, housing association leaders from across the UK gathered in Birmingham for Crimson's Building an AI-ready housing association leadership exchange. Hosted by Jordan Wheat, Housing Industry Practice Director at Crimson, the event brought together Digital Leaders to explore a challenge many social housing providers are now facing: how to move beyond AI experimentation and deliver meaningful outcomes for residents, colleagues and the wider organisation.
The sessions were led by Ollie Sinclair, CTO at Crimson, and Ian Bobbett, Chief Data Officer at Crimson, who combined practical demonstrations, sector research, and open discussion to explore where housing associations are on their AI journey, what is holding back progress, and what organisations should prioritise next.
What emerged was not a conversation about chasing the latest technology trend, but a discussion about how AI in social housing can deliver better resident outcomes through trusted data, strong governance and organisational readiness.
Research presented during the event, drawing on the National Housing Federation's How Housing Associations Are Adapting to AI report, published in 2026 and based on 2025 survey data, found that 47% of housing associations are already using AI in day-to-day operations. However, 87% reported low levels of AI knowledge, and 44% had no AI policy in place.
Further research from Service Insights and the University of Leeds, published in 2025, highlighted a similar governance and capability gap. While 31% of staff across ten housing associations were using AI in practice, only 13.5% were aware of an AI policy, 3.8% of an AI strategy, and 6.3% of AI training.
The live audience polling reflected the same picture. When attendees were asked where their organisation sat on its AI journey, 44% reported approved AI pilots, 44% reported live use cases, and 11% described their position as individual experimentation.
The debate is no longer whether AI has value. The challenge is moving from isolated use cases to governed, scalable and measurable outcomes. Many organisations are now focused on how to scale AI responsibly without losing control of governance, risk and business value.
Success starts with a clear AI strategy for social housing that aligns technology adoption with organisational outcomes rather than implementing AI in isolation.
A recurring theme throughout the event was that AI in housing is moving from curiosity to capability.
Many organisations are already experimenting with AI, but the organisations creating sustainable value are connecting AI to outcomes, trusted data and governance rather than deploying technology for its own sake.
For housing associations, this capability is emerging across practical areas such as tenant services, repairs intelligence, contact centres, compliance, assets and data management.
The value comes when these areas work together. Better service, better judgement, better prevention and better prioritisation all depend on trusted data, clear governance and a shared understanding of organisational outcomes.
The winners will not be the organisations with the most AI. They will be the organisations with the clearest outcomes, the strongest data foundations, and the highest levels of trust.
One of the most revealing findings came from the audience polling.
When asked what would make AI feel successful within the next 12 months:
For housing associations, the greatest value of AI is not in replacing people or systems, but in enabling:
Unlike many industries where AI discussions focus on productivity alone, housing leaders remain focused on improving services and outcomes for residents.
When attendees were asked about the biggest barrier to AI adoption:
Interestingly, none identified governance, system integration or leadership alignment as their primary challenge.
As Ian Bobbett explained during his session, organisations should resist the temptation to start with technology. The starting point should always be a clearly defined business problem, followed by the right combination of process change, analytics, automation and AI.
Service Insights and University of Leeds research, published in 2025, found that 93.1% of housing professionals believe good data quality underpins strategic goals. Yet only 43.2% find data easy to access, and just 42.2% trust its accuracy.
This was central to Ian Bobbett's message on AI readiness.
AI models are powerful, but they can only work with the information available to them. Poor-quality information will inevitably lead to poor-quality outcomes.
Becoming AI-ready is not about buying another tool. It starts with building trusted foundations.
Housing associations often rely on multiple housing association software platforms, including housing management systems, repairs platforms, asset management systems, CRM tools, contact centres, finance systems and document repositories. Bringing those sources together into a unified, governed data foundation is what enables better decision-making.
A practical example is Crimson's work with Great Places Housing Group, where a modern data platform helped provide more reliable reporting, improved insight and greater confidence in decision-making.
This reflects a wider shift across the sector towards creating trusted housing data and greater confidence in decision-making, rather than simply producing more reports
Technology platforms such as Microsoft Fabric, Purview, Microsoft 365 Copilot, Copilot Studio, and AI Foundry can support this journey, but the real value comes when trusted data, governance, and AI work together to improve services, compliance, and operational performance.
Housing associations making the greatest progress are taking an iterative approach: starting with a clearly defined use case, improving the data that supports it, establishing appropriate governance, and then scaling what works.
As AI adoption increases, establishing the right data management and governance foundations is critical.
Trust was another recurring theme throughout the day.
When attendees were asked how comfortable their organisation would be with AI supporting tenant-facing decisions:
The discussion evolved beyond the idea of a human in the loop to an expert in the loop.
For housing associations, AI can help surface information, identify risks and recommend actions. But responsibility for decisions affecting residents must remain with people.
The event highlighted several practical examples of AI in social housing already delivering value.
Research shared during the event included:
The lesson was consistent. The most successful examples of AI in housing are not large-scale transformations. They are practical solutions focused on genuine business challenges.
Ollie Sinclair's demonstration brought this opportunity to life.
By connecting data held across existing housing association software platforms, housing associations can move from reactive service delivery to proactive intervention
Examples included:
The most important takeaway was not the technology itself. It was the role of connected data in helping organisations identify risks earlier, support damp and mould complaint management, prioritise actions more effectively and improve outcomes for residents.
Want to see these use cases in action?
In this demonstration, Ollie Sinclair, CTO at Crimson, explores how housing associations can use connected data and AI to move from reactive service delivery to proactive intervention. The session covers complaint root cause analysis, predictive repairs, damp and mould complaint management, and regulatory insights, showing how trusted data can help organisations identify risks earlier and improve decision-making.
Watch: AI in Social Housing – Improving Repairs, Complaints and Compliance
The discussions also highlighted several challenges that repeatedly prevent AI initiatives from delivering value.
Starting with the tool instead of the outcome: Successful initiatives begin with a clearly defined business problem.
Building on poor data foundations: AI can only work with the information available to it.
Lacking a clear governance model: Housing associations need clear ownership, accountability and a practical approach to AI governance that enables innovation without introducing unnecessary risk.
Removing people from decision-making: Human oversight remains essential, particularly where resident outcomes are involved.
Failing to measure value: Organisations must be able to demonstrate improvements in service, efficiency, risk reduction or resident outcomes.
Start with the outcome. Strengthen your data. Put people and governance at the centre. Then measure the value to decide what should scale.
The organisations that will succeed with AI
The organisations that will realise the greatest value from AI in social housing are unlikely to be those with the flashiest demonstrations or the largest number of tools.
Successful AI adoption relies on governance, accountability, data quality, skills and value measurement as much as technology itself.
The organisations creating the greatest value from AI in social housing are connecting AI to clear outcomes, trusted data and strong governance.
They are:
In short, they are becoming data-ready before they become AI-ready.
Many housing associations already know where they want to get to with AI.
The more important question is whether the foundations are in place to get there.
Crimson's Data Assessment for Housing Associations helps organisations understand their current data maturity, identify opportunities, benchmark their readiness and create a practical roadmap for future AI adoption.
Before you can unlock the full potential of AI, you need confidence in the information that powers it.