Yes, we can replace civil servants with robots. The bigger question: ought we?

If an onshore public service process costs $1 to deliver and a nearshore one $0.60, the robotic equivalent might get your outlay per process down to as little as $0.15. But, challenged GlobalData thinker Jess Figueras at last week’s Think Smarter Working For Public Sector 2018 – should we actually go ahead and do that?

Posted 11 June 2018 at 8:42am by

The UK public sector has turned many staff jobs into more or less ones a robot could do…

So why not go the whole hog – and replace machine jobs with actual machines?

This was one of only one of the challenging thoughts leading public sector IT trend watcher Jessica Figueras confronted an audience of public sector tech practitioners and leaders with last week.

“The rise of automation will really change the nature of the way all this country’s 5.4m public service workers work,” the analyst told the audience at last Thursday’s successful one-day conference Think Smarter Working For Public Sector 2018.

As part of her keynote opening the event, Figueras also went on to detail how a quarter – £4.9bn – of the combined £18bn the UK spends on the public sector each year goes on provision of ICT tools to its own staff, a cost that could be considerably reduced by new approaches to provisioning and project delivery.

But it could also be reduced, she provocatively suggested, by a direct swap in jobs “where we already expect workers to work like machines, to deliver standardised results as quickly and cheaply as possible” to software that works 24×7, never takes a tea-break and won’t might well be “cheaper, faster and higher quality”.

Figueras also shared fresh GlobalData insights into the problem based on extensive research her team has just carried out across the UK public sector into reception of advanced technologies like AI and RPA (Robotic Process Automation).

The headline figure here: if an onshore process costs $1 and a nearshore one $0.60, the robotic equivalent might get your outlay per process down to as little as $0.15.

The ability to tailor services to specific needs is seen as the top investment driver, especially in health and higher education, the study suggests. Health sector respondents believe AI could have a positive impact on the treatment and therapy delivered to patients, while Higher Education respondents feel personalisation is of increasing importance as they respond to the challenges of marketisation.

The data, she told delegates, indicates a general willingness to invest in pilots and proof of concepts in the two technologies, but also a lack of skills and resources – meaning external advisory and support services will likely be crucial for helping to get projects off the ground.

But, she asked – even if we could replace many public sector processes with robots, should we?

“There are profound ethical considerations that we need to engage with here,” she warned, citing the recent Windrush scandal as an example of a process that “worked” but only because an essential element of humanity and ability to question case decisions by lower level staff had been removed from the system.

While some processes are definitely repeatable and predictable, she concluded, we must never lose sight of the fact that many are based on very complex human needs that even the best fuzzy logic simulator would struggle to cope with.

We’ll carry more reportage this week on the other findings of this very busy and content-rich conference – stay tuned.