- A quarter of the state’s IT systems are bound by legacy technology, the report claims
- £200 billion in AI-induced savings is being left on the table
- Modernization requires a dedicated task force and centralized procurement
Nearly a quarter of the 319 government IT systems being evaluated under the Legacy IT Risk Assessment Framework are ‘red-rated’, meaning not only that they use legacy technology and infrastructure but that they pose the highest operational risk of any system, new research has claimed.
However, government systems only provide a taste of how widespread legacy tech could be, with other public sector organizations such as the NHS and local police forces likely to have a higher percentage of ‘red-rated’ technologies – as much as 70% in some organisations, says a Re:State report.
However, quantifying the extent of legacy technology is not quite as easy as it seems, with 15% of public bodies admitting that they cannot provide an accurate overview or asset inventory of their legacy IT portfolio, highlighting major gaps across visibility.
Legacy tech is a burden in the public sector
To further add to the complexity, Re:State criticized public bodies and government agencies for fixing problems with temporary fixes.
Like band-aids, these fixes tend to be applied over existing and legacy systems, resulting in additional complexity on top of old code, making upgrading at a later date more difficult.
With older hardware particularly known for its inefficiency and higher energy consumption, there are both environmental and fiscal costs to not upgrading.
Financial modeling by the Department for Science, Innovation and Technology (DSIT) shows that this failure to upgrade could cost the UK an estimated £45 billion annually in both lost productivity and unrealized savings.
The report even suggests that keeping existing legacy systems functional accounts for about half of the government’s IT spending.
Adding artificial intelligence on top of legacy systems is not the answer
While artificial intelligence promises to bring major productivity improvements to administrative workloads, implementation is not without its own challenges.
Fragmented technology, poor data quality and complex software layers stand in the way of a full government AI rollout, despite the Tony Blair Institute predicting the government could realize £200 billion in AI-related savings over the next five years.
However, Re:State does not shift all the blame onto public bodies and public authorities. The report also acknowledges that many departments routinely extend contracts as a result of commercial models, which can induce vendor lock-in or make migration technically challenging or expensive.
The 50-year-old Police National Computer (PNC) was cited as a key example – it has been in place since 1974 and now houses more than five decades of highly sensitive information.
The fact that government agencies must also outsource software contracts allows private companies to dictate their own terms, says Re:State.
Digital modernization is a long-term goal
Although it is not a task to address the challenges overnight, the research requires the establishment of a digital modernization taskforce, similar to the recent Vaccine Taskforce, which will work across departments and collaborate directly with the relevant ministers.
Mandatory Technology Impact Assessments (TIAs) are also recommended, whereby logistical feasibility, life cycle costs, legacy dependencies and cyber security risks should be discussed with ministers prior to technical changes.
There are also advantages to a centralized procurement system, like centralized procurement under the US General Services Administration (GSA), designed to drive procurement costs down under the Trump administration.
“Technology will not stop changing, and no government will ever end modernization,” the paper concludes. “But aging systems need not inevitably become a source of persistent state weakness.
Follow TechRadar on Google News and add us as a preferred source to get our expert news, reviews and opinions in your feeds.
