7 ways digital assistants and AI will help transform public services

Profile picture for user Mark Gibbison By Mark Gibbison January 11, 2017
Digital assistants and artificial intelligence could transform public services. Unit4's Mark Gibbison sets out seven ways they will help

Man walking from building using digital tablet and mobile phone © s4svisuals - Fotolia.com
Globally, there is increasing pressure on large public institutions to become more efficient. With greater transparency than ever before, taxes need to be spent wisely if politicians are to maintain favour with the voting public. One key challenge is the trend for long-term growth of the state combined with the increased demand for services driven by an ageing population.

Just employing more workers to meet this demand would be a short-term strategy. Not only is there a general picture of a decline in funding, at the same time the global resource pool is shrinking. Analysis by McKinsey’s Global Institute suggests the growth in population is slowing and life expectancy is increasing; the working age population will therefore shrink and reduce the human capacity to deliver public services.

The answer to doing more with less, as it often is, is to utilise technology. In this article I want to outline why artificial intelligence has the greatest potential of all existing and previous technologies to manage the delivery of public services. Here are some potential use cases:

1. Serving citizens

Online digital assistants will serve the information that citizens require, 24x7x365. They won’t require a salary, they won’t take sick leave and they have the computational power to deliver almost instant answers extrapolated from terabytes of big data. What’s more, they will have the capability to learn patterns from the wider population and individual citizens to personalise and therefore improve the services.

2. Developing proactive strategy

Public services are complex and changeable. Keeping track of all nationwide project successes (or failures) across millions of data points is impossible. Naturally this means mistakes are often repeated and work duplicated. Digital assistants, however, have the capability to learn what approaches succeed and proactively suggest and prebuild approaches. Take for instance the construction of a new hospital. Digital assistants could recognise the project aims and provide a suggested plan based on previous successful endeavours. This would save the time of building a project plan and provide significant insight to make it a success.

3. Updating records

Citizens and employees are prompted at regular intervals to confirm their personal data to make sure it is up-to-date. Digital assistants have the capability to revolutionise stakeholder interaction with public bodies. Through automation, bots will significantly reduce the administration and research citizens and employees must undertake with the multiple touch points they have with the state. The right information can be requested at the right time by digital assistants, or citizens can more easily access their records online using natural language requests.

4. Improving value delivered

When citizens or employees wish to know which services are available to them given their personal circumstances, the vastness of departments and their sheer number, makes this task complex and daunting, especially for the elderly and disadvantaged.

Digital assistants could help by suggesting to employees and citizens the services that would have the biggest impact on individuals. The digital assistants would provide advice based on parameters such as budgets, severity of each citizen’s needs and entitlement.

5. Identifying fraud

Existing fraud detection systems flag dubious-looking financial dealings according to a set of rules, such as challenging employee expense claims in excess of a fixed cash amount or those that are five percent higher than claims submitted by peers in similar positions. While these methods can help to stop a number of fraudulent expense claims, they are only as good as the rules you code. But artificial intelligence can take fraud identification to the next level.

Exposing machine learning systems to new data allows them to learn and spot patterns of behaviour and, crucially, anomalies in those patterns.

For a sector that processes billions of financial transactions a year from benefits, to tax, to grants, to supplier payments and more, the potential savings by identifying the fraudulent minority are considerable.

6. Managing uncertainty

Existing legacy systems in public services are robust, they can carry out repeatable actions with certainty. But public services are anything but certain, and that is where artificial intelligence can make a huge difference. The cognitive processing power of artificial intelligence enables for adaptation and the suggestion of strategies based on past successes and failures held within big data. It will also allow for the automation of many administrative processes hitherto dealt with by manual intervention. Already technology is available that can automate monthly expenses based upon an individual’s travel using GPS on their phone, photographs of receipts and historical expense claims.

7. Improving productivity

Right now, across sectors, on average workers spend up to 54% of their time on tasks that are not core to their roles, such as administration or travelling. In the future digital assistants will be able to manage increasingly complex administrative tasks faster than humans, in real time – vastly increasing the productivity of back office operations and the service delivered to all stakeholders.

This does not just mean more productive administration; it also means more efficient, wider operation. Employees, a finite resource, will have considerably more time to focus on delivering crucial frontline and citizen-centric services such as healthcare, infrastructure management and education.