Housing Technology interviewed AI practitioners from DASH (Demystifying AI for Social Housing), Infinity Group, Mobysoft, NEC Housing, Plentific and TrustMarque on the role of artificial intelligence (AI) in housing, including the benefits to housing staff and tenants, pitfalls to avoid, how to get started, cutting through the hype, real-life examples and ethics.
The roles of AI, RPA, ML & NLP
Sarah McRow, head of housing sales at Infinity Group, said, “Think of artificial intelligence (AI), robotic process automation (RPA), machine learning (ML) and natural language processing (NLP) like a team.
“RPA handles the repetitive chores, ML quietly watches for patterns, NLP listens and talks, and AI weaves them together, turning your raw operations into smarter, kinder services.”
Emre Kazan, co-founder and chief technology officer at Plentific, said, “AI drives smarter decisions, RPA bridges the integration gaps between legacy systems by mimicking human tasks, ML predicts future trends and NLP improves communication, together transforming housing operations into seamless, proactive services.
“Where APIs fall short, RPA steps in; where rules end, ML begins; where data speaks, NLP listens. AI unites them to deliver proactive, data-driven housing services.”
Professor Alan Brown, AI expert at DASH (Demystifying AI for Social Housing) and professor in the digital economy at University of Exeter Business School, said, “Each plays a distinct role in building a more intelligent housing system. RPA automates repetitive tasks, such as logging repairs and generating compliance reminders. ML identifies patterns in, say, arrears or disrepair in order to flag potential problems early, and NLP powers tenant-facing tools such as chatbots, improving access and responsiveness.
“AI pulls these tools together to support better, faster decisions. At North Star Housing, for example, AI-driven automation means that 1000s of safety actions are now completed on time.”
Realistic housing use-cases
Chris Fleck, chief technology and product officer at Mobysoft, said, “AI is already delivering tangible value within income collections, repairs management and customer service. Our RentSense customers, for example, have seen their arrears reduce to a much greater extent than housing providers not using our AI-powered platform.
“Other real-life examples include the automated triaging of tenant communications, predictive modelling for voids and re-lets, and smart resource allocation based on likely demand.”
Trevor Hampton, director of housing solutions at NEC Housing, said, “AI is helping housing providers to predict which tenants are at risk of missing their rent payments by identifying patterns of circumstances, such as the likelihood of debt or experiencing financial hardship, and then intervene early to improve outcomes for both the resident and the housing provider. ML models can identify the combined factors which could cause vulnerability, such as someone who has recently been in hospital, lives in an area with antisocial behaviour or is socially isolated following a bereavement. And for asset management, AI can predict which properties are likely to have damp and mould by tracking contributory factors, such as financial hardship, over-occupation and poor ventilation.
“There are also new opportunities for housing providers to manage their repairs using a combination of NLP or, more specifically, a large-language model (LLM). If a resident takes a photograph of a leaking tap, the housing provider could use NLP to describe the photograph and say if the leak is from the faucet, the tap itself or a flange below the tap, thereby helping their maintenance teams to plan repairs and arrive at the property with the right skills and equipment.”
Seb Burrell, head of AI at Trustmarque Group, said, “AI can be used to streamline maintenance requests, predict property problems before they escalate and improve tenant communications through chatbots. These applications can all help in reducing response times and enhancing service delivery. Additionally, AI can help in analysing patterns in rent payment to identify tenants who might need support, allowing for better-targeted interventions.”
Is AI under- or over-hyped?
Infinity’s McRow said, “AI in housing is both over-hyped and under-imagined. It’s oversold as a miracle fix for deep social problems, yet undersold as a practical tool for giving housing staff the necessary breathing space to do what only humans can do – listen, care and build trust.
“AI isn’t a silver bullet; it’s a practical tool that’s often underestimated. Used thoughtfully, it can give staff more time to focus on building trust and making a real difference to tenants’ lives.”
Plentific’s Kazan said, “Forward-thinking organisations recognise the importance of a structured, high-quality data foundation and a well-architected software infrastructure as the critical foundations without which AI can’t function reliably, reach its full potential, be trusted to support operational decision-making or deliver consistent outcomes at scale.
“AI is both an opportunity and an imperative. The real challenge for housing providers isn’t deciding whether to adopt AI, it’s how swiftly and effectively they can build the underlying data and software foundations that make AI truly valuable.
“When records and operational data are spread across PDFs, spreadsheets and emails, it becomes difficult for AI models to generate accurate or timely insights. It’s tempting to believe that all unstructured data, emails, reports and raw logs can simply be collected in a data lake for AI agents to interpret later. Although AI has made impressive progress in processing unstructured information, relying on a ‘data swamp’ approach remains inefficient; AI performs best when it has access to well-structured, contextualised data, rather than needing to decipher meaning from an organised collection of documents.”
DASH’s Brown said, “AI is over-hyped when it’s positioned as a quick fix for complex, human systems. But it’s under-hyped when we look at the real improvements already emerging.
“For example, ExtraCare Charitable Trust’s use of AI to support independent living is subtle but powerful, and Manningham Housing’s chatbot delivers better service, not through flashy technology, but by solving a persistent contact challenge. AI’s value isn’t in transformational headlines, it’s in quiet, measurable progress.”
Business risks and considerations
Mobysoft’s Fleck said, “The key risks are data quality, transparency and an over-reliance on automation. An AI is only as good as the data it’s fed so ensuring high-quality, well-labelled and representative data is essential.
“There’s also the governance aspect. Housing providers need to be clear on when and how decisions are being made or supported by ‘responsible’ AI, particularly when those decisions affect tenants. For example, Mobysoft designs responsible AI tools to augment, but not replace, human decision-making in order give housing staff clearer insights rather than removing accountability.”
NEC Housing’s Hampton said, “There are risks associated with using generic AI and LLMs which haven’t been properly trained with housing data. AI is only as good as the data used to train it so it’s important to maintain high standards of data management and to tackle the root cause of any problems with data quality.
“Users of AI need to understand what a model is doing, such as why it is predicting that a resident might be vulnerable or how it identifies a property with damp and mould. To avoid using ‘black box’ technology, housing providers should adopt ‘explainable AI’, with clear reasoning behind the AI’s recommendations.
“It’s good practice to examine the probability that an AI-generated recommendation is accurate. By saying we are 99 per cent confident that a particular house has damp and mould for these reasons or 85 per cent confident that a specific tenant will fall into debt owing to these factors, we reduce the risk of relying too heavily on AI to make the final call.”
Plentific’s Kazan said, “A common risk is underestimating the importance of a structured software and data foundation. Without clean, connected and well-architected systems, AI struggles to deliver insights that are timely, reliable or actionable. Without that foundation, even the most advanced AI models can fall short of their potential and undermine trust in AI-supported decision-making.
“Housing providers must carefully manage their AI implementation risks, including ensuring data privacy, addressing potential biases within AI algorithms, avoiding an over-reliance on automated systems without sufficient human oversight and complying with relevant regulatory frameworks.”
Ethical considerations
Trustmarque’s Burrell said, “Ethically, housing providers must ensure that their AI applications don’t inadvertently discriminate against any group and that any AI-derived decisions AI are fair and justifiable. Transparency is key, so tenants should be informed when AI is used in processes that affect them and there should be avenues for human review and appeal.”
Mobysoft’s Fleck said, “Fairness, accountability and transparency must be at the heart of any use of AI. Algorithms mustn’t reinforce biases or exclude vulnerable customers, and housing providers must ensure that their AI models are explainable, auditable and subject to human oversight.
“We believe it’s ethical and essential to use AI to improve service delivery, reduce arrears and prevent problems such as disrepair, but it must be balanced with privacy, consent and a human touch.”
NEC Housing’s Hampton said, “The key to ethical AI is to make sure that decision-making remains in the hands of humans, not machines. For example, if an AI model recommends contacting a person to ask if they are in debt, a human should first look at why the recommendation has been made.”
Where are the ‘quick wins’?
Infinity’s McRow said, “The best place to start is where AI can take pressure off your teams, such as spotting maintenance problems early, triaging routine questions or flagging tenancy risks. It’s about shining a light on hidden problems, not replacing human connections, so use AI to illuminate hidden problems such as unnoticed complaints about mould, maintenance backlogs and patterns of missed communications. Use AI to shine a light, not to cast shadows.”
Plentific’s Kazan said, “AI is a powerful tool but it’s not a ‘one size fits all’ solution. Its suitability depends on the nature of the problem, the quality of the available data and the clarity of the desired outcomes. In some instances, traditional software logic or process redesign may deliver better results with less complexity so it’s important to assess each case on its own merits before deciding whether AI is the right approach.
“Some quick wins include automating straightforward administrative tasks, deploying AI chatbots to handle common tenant enquiries and leveraging AI and ML for predictive maintenance.
“Start small, solve real problems and then scale from your initial success because AI adoption works best when it’s grounded in everyday results. Furthermore, most software providers are implementing AI within their existing products so take advantage of these ready-made capabilities rather than going into a ‘green field’ AI implementation.”
NEC Housing’s Hampton said, “A good starting point would be to predict which tenants are likely to be experiencing financial hardship and have difficulty paying their rent. This is one of the more straightforward uses of AI because it’s very structured and mathematical, with a focus on finance-based data.
“Using AI to manage properties could then be your next step, although making predictions about housing assets tends to be more complex because of the numerous variables you need to consider, such as building archetype, construction details, building materials and surveys.”
AI disclosure
Infinity’s McRow said, “Housing providers should definitely disclose their use of AI but not just buried in a 300-page privacy policy – tenants deserve to know if a chatbot, repairs system or a housing waiting-list is AI-powered.
“It’s important to be open. Tenants should know when AI is being used and always have the option to speak to a person if they need to. Being clear and honest helps build trust and shows that technology is there to support, not replace, human relationships.”
Trustmarque’s Burrell said, “Transparency about the use of AI is crucial. Housing providers should inform tenants about AI applications through clear communication channels, such as tenant portals or informational brochures, explaining how AI is being used and the benefits it brings, while also providing human contact points for questions or concerns.”
DASH’s Brown said, “Housing provider should be open about their use of AI because tenants deserve to know when a chatbot is answering their query or when a predictive system is flagging their account. From our experience, Manningham Housing’s decision to be open about AI helped to build confidence and not confusion.
“To that end, AI disclosure should be integrated into tenant communications, from welcome packs to online portals and regularly reviewed with scrutiny groups. Being open isn’t just responsible, it’s good business.”
Gains for housing staff
Mobysoft’s Fleck said, “For staff, AI reduces manual work, highlights the right priorities and improves job satisfaction. For example, housing officers can spend more time supporting their tenants and less time trawling through caseloads or repair logs.”
NEC Housing’s Hampton said, “AI can carry out tasks and identify patterns from complex data in a fraction of the time it would take a human (or even a team of humans). AI can predict, say, vulnerabilities or risks of debt, and housing staff can take that valuable information and use it with a human touch.”
Plentific’s Kazan said, “AI helps housing staff by automating their routine tasks and reducing human error, freeing them to focus on strategic priorities and, importantly, fostering stronger relationships with tenants.”
Benefits to tenants
Infinity’s McRow said, “For tenants, the benefits of AI are felt through faster communications, quicker resolutions of their problems and services that feel more connected and responsive. Behind the scenes, AI strengthens data quality, helping housing providers to manage their homes more securely and reliably, all of which builds better experiences and greater satisfaction for the people they serve.”
Trustmarque’s Burrell said, “For tenants, AI should lead to faster responses, more accurate information and improved service overall, enhancing their satisfaction and trust in their housing provider.”
DASH’s Brown said, “AI can result in faster repairs, clearer updates and better support. For example, ExtraCare uses AI to support its older residents in maintaining their independence, a ‘win’ for both well-being and service efficiency. When done right, AI doesn’t replace people, it empowers them.”
Sector-wide examples
NEC Housing’s Hampton said, “Wolverhampton Homes, PA Housing, Lambeth Council and Gateshead Council are each using sector-specific AI to make quantitative predictions about which of their assets are most at risk of damp and mould.
“For example, Southwark Council is using AI tools to analyse historical data on how its tenants manage their finances and pay their rent. The AI takes into account multiple factors and provides a risk score, helping the council to spot people at risk of missing payments in the future, even if they appear to be managing at the moment.”
DASH’s Brown said, “We’ve found a striking breadth of success across very different types of organisations. These include Together Housing which has developed an AI model capable of identifying tenants at high risk of tenancy failure with above 80 per cent accuracy, and Incommunities’ pilot project for predictive maintenance, in collaboration with University of Bradford, has already shown its potential to reduce the need for emergency repairs.
“As another example, Manningham Housing deployed an AI-powered chatbot to provide tenants with accurate, 24/7 responses to common queries. This not only reduced pressure on housing staff but also improved ‘first-time fix’ rates.
“Together Housing, Incommunities and Manningham Housing each demonstrate that AI innovation isn’t limited by size or scale.”
Infinity’s McRow said, “North Star Housing’s legacy systems have been replaced with our BRIKHousing software for a modern, AI-driven approach to compliance and asset management. Using AI to scan and check certifications, covering everything from gas safety and fire maintenance to lift inspections and legionella testing, has transformed how North Star’s compliance team works.
“Routine processing that previously took around 25 hours per week has been reduced to just a couple of days, giving North Star’s team time to focus on new initiatives such as damp, mould and condensation monitoring.”
Housing Technology would like to thank Prof. Alan Brown (DASH), Sarah McRow (Infinity Group), Chris Fleck (Mobysoft), Trevor Hampton (NEC Housing), Emre Kazan (Plentific) and Seb Burrell (Trustmarque Group) for their editorial contributions to this article.
