Social housing is in the midst of a customer service revolution. We have seen it in many ways already, with the development of customer service centres, web sites, online payments and social media. As a consequence, over recent years we have seen the loss of rent collectors, repairs inspectors, local offices and resident wardens. More recently, there has been the growth of online self-service, which is likely to have the most dramatic impact on services over the next five years. That’s why the development and introduction of artificial intelligence (AI) in customer services is significant. It moves things forward, providing a system to empower tenants to deal with their own queries or service requests.
The application of AI for customer services results from improvements in automated voice recognition. AI can help customer service providers to more fully exploit the power that can be derived from sets of data. AI does not degrade over time, unlike other facets of production, but rather assists societies to operate and enhances how data about people is collated and interrogated. Moreover, with developments in AI, computers have the ability to learn from their interactions.
So, are we seeing the beginning of the end of human domination? It was forecast that by 2025, the computer would be able to out-perform a human. That now looks like being well over-taken. Writing in The Guardian, Ian Sample explained how deep learning via artificial general intelligence is a step closer, with Google’s DeepMind programme which can develop skills to achieve one task after another. Uniquely, DeepMind does not forget how it solved past problems and can use this acquired knowledge to tackle new issues. Sequential learning and the ability to remember old skills and apply them to tasks comes naturally to humans.
DeepMind is still not capable of the general intelligence that humans use when facing new challenges. Writing in the Financial Times, John Thornhill said, “If we ever reach the point of a technological singularity, when computers outsmart humans, productivity growth will accelerate exponentially. From that moment, a computer superintelligence will rapidly discover everything left to discover. This ‘master algorithm’ will be the last invention that man makes. It will be able to derive all knowledge in the world – past, present and future – from data.” The ultimate aim is for AI machines which match human intelligence and we are not too far away from this.
A study by McKinsey suggested that 45 per cent of US employees’ work time is spent on tasks that could be automated with existing technologies. A recent report from Accenture and Frontier Economics suggested that AI-enabled technologies could double the economic growth rates of many advanced countries by 2035. There can be no doubt therefore that AI will enable many roles in housing, especially in customer services, to be automated.
Last year, a computer using AI beat the world champion at the incredibly complex game of Go. What was significant, and possibly worrying, was that the computer made unpredictable moves that ‘surprised’ its developers. This betrays a lack of control inherent in the design, which may be OK with a game such as Go, but raises important ethical and governance issues.
How do we make sure the machines we ‘train’ don’t perpetuate and amplify the same human biases that plague society? In a utilitarian way, do we programme machines to maximise the happiness of the greatest number of people? How does this affect those who don’t fit the mould? How do we control what we have started?
In the final analysis, an expert is required to initially give the system the knowledge and design the flow pathways. Interestingly, that is exactly how Omfax’s Keyfax works; scripts are developed to initially reflect the known and repetitive pathways for a range of enquiries. They are then refined and extended, based on feedback. Experts are called in to contribute their knowledge and understanding to help design the scripts’ pathways. Data sets relating to the tenant, their property, their relationship with the organisation and their history are then added to create intelligent, personalised responses.
There is power in the Keyfax system, which is so much more than a repairs diagnostic tool, just as AI aims to fully exploit the power of data sets. Keyfax assists and trains customers to self-diagnose in an aspirational way, much akin to artificial intelligence. AI is pushing at the boundaries of customer services in social housing and may, in the not too distant future, make current technologies obsolete, as the sector continues to innovate. Keyfax has already made strides in that direction.
Peter Graddon is a director of Omfax Systems.