Technology
How to transform public services and eliminate IT debt

Paul Liptrot, Partner – UK Government & Public Sector at Kyndryl
Any organisation embarking on a digital transformation project will know that there are significant challenges to overcome. While the benefits at the end are impossible to ignore, the barriers to success are complex and wide ranging. Add in legacy public sector applications to the process, and you open up a whole new layer of complexity.
Whether it’s cost-cutting pressures or competing interests holding back the process, product development and delivery can be significantly harder than it is in the private sector. Not just that, but across Whitehall it is broadly recognised that there’s a significant amount of technical debt to deal with too, meaning many Departments are typically starting from a place far behind that of their commercial counterparts.
Technical debt has been acquired in the public sector over several years (even decades), and for many reasons. Predominantly, ongoing budget constraints and loss of resources mean it has been easy to put off long-term, labour-intensive maintenance and modernisation projects, in favour of feel-good ‘quick wins’ that can be delivered quickly and the impact felt fast.
Furthermore, legacy, proprietary platforms that many public sector systems are developed and operating on, have undergone significant customisation and bespoke development over many years. Not only do these support mission-critical processes, making it hard to get downtime for improvements approved, but they are notoriously tricky (read: time-consuming) to unpick and migrate to newer, more modern platforms and infrastructures.
Too often, both public and private sector organisations have had a superficial opinion of what it takes to address tech debt, focusing on dealing with the often out-of-support components rather than the underlying problems. As a result, tech debt is only getting worse, reducing workforce productivity, inhibiting innovation and keeping public services stagnant through a lack of agility.
For public sector organisations looking to address the problem, the first step must be to assess and understand which services are going to be needed to deliver current policy and customer outcome objectives. With this clarity organisations can then design their Target Operate Model (TOM) to not only support these but also put in place processes and technologies that will be agile enough to adopt and adapt for the future. Understanding the TOM will then allow CXOs and their teams to make informed decisions on which approaches to take, which technologies to retain, which to replace and which to adopt public and private cloud services for in an overall hybrid environment.
With a TOM in place, priorities can be established, required investments can be understood and organisation teams and squad structures can be established within a governed program to remove tech debt. Informed decisions can be taken per business service or application with options typically including:
Retire and start new
Evaluate whether it is even worth modernising the underlying technologies. Would it be easier or more beneficial to instead migrate the business processes and data to something else? This could be a modern SaaS platform or a lighter-weight, micro-service-based replacement application. Where suitable SaaS products exist, moving legacy services to these transfers the risks and costs of developing and maintaining underlying application software and infrastructure to the SaaS provider.
Re-platform
Where SaaS isn’t an option, building products on PaaS can reduce the risk of tech debt by moving the responsibilities for maintaining, patching and updating the underlying infrastructure, operating systems and middleware to the cloud provider. As PaaS migrations involve refactoring legacy applications to fit the platform offered by a service provider, it can be time consuming. However, once it’s complete, it takes the maintenance time away from programmers, freeing them up to deploy new applications faster.
“Lift and shift”
One of the easier and least expensive ways to migrate an existing workload to the cloud is an IaaS migration, sometimes called a “lift and shift”. That’s because you move it in its current form, with minimal changes, and run it on cloud-native resources instead. This is particularly useful where the technical debt resides in hosting locations and physical hardware, but typically isn’t the answer where the debt relates to applications or software. In these cases, a lift and shift can be useful as part of a longer term program, by buying more time to modernise the applications once into Public Cloud and freed from the immediate risks associated with ageing hardware or building closures.
Modernise in-situ
You could choose to slowly migrate a legacy system by replacing specific components over a period of time. There are many variants to this approach but one of the most common is the “Strangler Fig” pattern which involves creating a parallel new landing zone and slowly redirecting from the existing application components to the new ones as replacement functionality and services are implemented. Eventually, the legacy is retired completely.
Whichever methods are adopted, once tech debt has been eliminated, that’s not the end of the road. Maintaining the TOM is an ongoing monitoring exercise to ensure that it doesn’t begin to accumulate again. Here are six top tips to avoid it recuring:
- Run teams that maintain and constantly update products rather than big periodic projects. The ongoing level of investment may seem high but the TCO will be lower, with the product itself remaining evergreen, supporting innovation and avoiding build-up of tech debt.
- Maintain tech roadmaps and review regularly, with tech refresh built into investment cases up front.
- Implement modern DevSecOps and Infrastructure as Code methodologies to ensure that services can be updated and re-deployed easily.
- Consider options like Low-code/No-code and RPA for building applications and workflows where these are suitable, avoiding the need to write and maintain code.
- Design for micro-services over monoliths. This allows parts of applications to be updated and modernised in isolation from the rest of the service. Principles of abstraction help here, as well as designing for components/services to be joined up using APIs and loose-coupling.
- Feeling overwhelmed? Engage a partner you trust that can help advise, build and, if required, operate your technology transformation.
You may like
Business
Hype, Hysteria & Hope: AI’s Evolutionary Journey and What it Means for Financial Services

Source: Finance Derivative
Written by Gabriel Hopkins, Chief Product Officer at Ripjar
Almost a year to the day since ChatGPT launched, the hype, hysteria, and hope around the technology shows little signs of abating. In recent weeks OpenAI chief Sam Altman was removed from his position, only to return some days later. Rishi Sunak hosted world leaders at the UK’s AI Safety Summit, interviewing the likes of Elon Musk in front of an assembly of world leaders and tech entrepreneurs. While behind the scenes, AI researchers are rumoured to be close to even more breakthroughs within weeks.
What does it all mean for those industries that want to benefit from AI but are unsure of the risks?
It’s possible that some forms of machine learning – what we used to call AI – have been around for a century. Since the early 1990s, those tools have been a key operational element of some banking, government, and corporate processes, while being notably absent from others.
So why the uneven adoption? Generally, that has been related to risk. For instance, AI tools are great for tasks like fraud detection. It’s a well-established that an algorithm can do things that analysts simply can’t by reviewing vast swathes of data in milliseconds. And that has become the norm, particularly because it is not essential to understand each and every decision in detail.
Other processes have been more resistant to change. Usually, that’s not because an algorithm couldn’t do better, but rather because – in areas such as credit scoring or money laundering detection – the potential for unexpected biases to creep in is unacceptable. That is particularly acute in credit scoring when a loan or mortgage can be declined due to non-financial characteristics.
While the adoption of older AI techniques has been progressing year after year, the arrival of Generative AI, characterised by ChatGPT, has changed everything. The potential for the new models – both good and bad – is huge, and commentary has divided accordingly. What is clear is that no organisation wants to miss out on the upside. Despite the talk about Generative and Frontier models, 2023 has been brimming with excitement about the revolution ahead.
Two Objectives
A primary use case for AI in the financial crime space is to detect and prevent fraudulent and criminal activity. Efforts are generally concentrated around two similar but different objectives. These are thwarting fraudulent activity – stopping you or your relative from getting defrauded – and adhering to existing regulatory guidelines to support anti-money laundering (AML), and combatting the financing of terrorism (CFT).
Historically, AI deployment in the AML and CFT areas has faced concerns about potentially overlooking critical instances compared to traditional rule-based methods. Within the past decade, and other regulators initiated a shift by encouraging innovation to help with AML and CFT cases. Despite the use of machine learning models in fraud prevention over the past decades, adoption in AML/CFT has been much slower with a prevalence for headlines and predications over actual action. The advent of Generative AI looks likely to change that equation dramatically.
One bright spot for AI in compliance over the last 5 years, has been in customer and counterparty screening, particularly when it comes to the vast quantities of data involved in high-quality Adverse Media (aka Negative News) screening where organisations look for the early signs of risk in the news media to protect themselves from potential issues.
The nature of high-volume screening against billions of unstructured documents has meant that the advantages of machine learning and artificial intelligence far outweigh the risks and enable organisations to undertake checks which would simply not be possible otherwise.
Now banks and other organisations want to go a stage further. As Generation AI models start to approach AGI (Artificial General Intelligence) where they can routinely outperform human analysts, the question is when, and not if, they can use the technology to better support decisions and potentially even make the decisions unilaterally.
AI Safety in Compliance
The 2023 AI Safety Summit was a significant milestone in acknowledging the importance of AI. The Summit resulted in 28 countries signing a declaration to continue meetings to address AI risks. The event led to the inauguration of the AI Safety Institute, which will contribute to future research and collaboration to ensure its safety.
Though there are advantages to having an international focus on the AI conversation, the GPT transformer models were the primary focus areas during the Summit. This poses a risk of oversimplifying or confusing the broader AI spectrum for unaccustomed individuals. There is a broad range of AI technologies with hugely varying characteristics. Regulators and others need to understand that complexity. Banks, government agencies, and global companies must exert a thoughtful approach to AI utilisation. They must emphasise its safe, careful, and explainable use when leveraged inside and outside of compliance frameworks.
The Road Ahead
The compliance landscape demands a review of standards for responsible AI use. It is essential to establish best practices and clear objectives to help steer organisations away from hastily assembled AI solutions that compromise accuracy. Accuracy, reliability, and innovation are equally important to mitigate fabrication or potential misinformation.
Within the banking sector, AI is being used to support compliance analysts already struggling with time constraints and growing regulatory responsibilities. AI can significantly aid teams by automating mundane tasks, augmenting decision-making processes, and enhancing fraud detection.
The UK can benefit from the latest opportunity. We should cultivate an innovation ecosystem with is receptive to AI innovation across fintech, regtech, and beyond. Clarity from government and thought leaders on AI tailored to practical implementations in the industry is key. We must also be open to welcoming new graduates from the growing global talent pool for AI to fortify the country’s position in pioneering AI-driven solutions and integrating them seamlessly. Amid industry change, prioritising and backing responsible AI deployment is crucial for the successful ongoing battle against all aspects of financial crime.
Business
Using AI to support positive outcomes in alternative provision

By Fleur Sexton
Fleur Sexton, Deputy Lieutenant West Midlands and CEO of dynamic training provider, PET-Xi, with a reputation for success with the hardest to reach,
discusses using AI to support excluded pupils in alternative provision (AP)
Exclusion from school is often life-changing for the majority of vulnerable and disadvantaged young people who enter alternative provision (AP). Many face a bleak future, with just 4% of excluded pupils achieving a pass in English and maths GCSEs, and 50% becoming ‘not in education, employment or training’ (NEET) post-16.
Often labelled ‘the pipeline to prison’, statistics gathered from prison inmates are undeniably convincing: 42% of prisoners were expelled or permanently excluded from school; 59% truanted; 47% of those entering prison have no school qualifications. With a prison service already in crisis, providing children with the ‘right support, right place, right time’, is not just an ethical response, it makes sound financial sense. Let’s invest in education, rather than incarceration.
‘Persistent disruptive behaviour’ – the most commonly cited reason for temporary or permanent exclusion from mainstream education – often results from unmet or undiagnosed special educational needs (SEN) or social, emotional and mental health (SEMH) needs. These pupils find themselves unable to cope in a mainstream environment, which impacts their mental health and personal wellbeing, and their abilities to engage in a positive way with the curriculum and the challenges of school routine. A multitude of factors all adding to their feelings of frustration and failure.
Between 2021/22 and 2022/23, councils across the country recorded a 61% rise in school exclusions, with overall exclusion figures rising by 50% compared to 2018/19. The latest statistics from the Department for Education (DfE), show pupils with autism in England are nearly three times as likely to be suspended than their neurotypical peers. With 82% of young people in state-funded alternative provision (AP) with identified special educational needs (SEN) and social emotional and mental health (SEMH) needs, for many it is their last chance of gaining an education that is every child’s right.
The Department for Education’s (DfE) SEND and AP Improvement Plan (March 2023).reported, ‘82% of children and young people in state-place funded alternative provision have identified special educational needs (SEN) 2, and it (AP) is increasingly being used to supplement local SEND systems…’
Some pupils on waiting lists for AP placements have access to online lessons or tutors, others are simply at home, and not receiving an education. In oversubscribed AP settings, class sizes have had to be increased to accommodate demand, raising the pupil:teacher ratio, and decreasing the levels of support individuals receive. Other unregulated settings provide questionable educational advantage to attendees.
AI can help redress the balance and help provide effective AP. The first challenge for teachers in AP is to engage these young people back into learning. If the content of the curriculum used holds no relevance for a child already struggling to learn, the task becomes even more difficult. As adults we rarely engage with subjects that do not hold our interest – but often expect children to do so.
Using context that pupils recognise and relate to – making learning integral to the real world and more specifically, to their reality, provides a way in. A persuasive essay about school uniforms, may fire the debate for a successful learner, but it is probably not going to be a hot topic for a child struggling with a chaotic or dysfunctional home life. If that child is dealing with high levels of adversity – being a carer for a relative, keeping the household going, dealing with pressure to join local gangs, being coerced into couriering drugs and weapons around the neighbourhood – school uniform does not hold sway. It has little connection to their life.
Asking the group about the subjects they feel strongly about, or responding to local news stories from their neighbourhoods, and using these to create tasks, will provide a more enticing hook to pique their interest. After all, in many situations, the subject of a task is just the ‘hanger’ for the skills they need to learn – in this case, the elements of creating a persuasive piece, communicating perspectives and points of view.
Using AI, teachers have the capacity to provide this individualised content and personalised instruction and feedback, supporting learners by addressing their needs and ‘scaffolding’ their learning through adaptive teaching.
If the learner is having difficulty grasping a concept – especially an abstract one – AI can quickly produce several relevant analogies to help illustrate and explain. It can also be used to develop interactive learning modules, so the learner has more control and ownership over their learning. When engaged with their learning, pupils begin to build skills, increasing their confidence and commitment.
Identifying and discussing these skills and attitudes towards learning, with the pupil reflecting on how they learn and the ways they learn best, also gives them more agency and autonomy, thinking metacognitively.
Gaps in learning are often the cause of confusion, misunderstandings and misconceptions. If a child has been absent from school they may miss crucial concepts that form the building blocks to more complex ideas later in their school career. Without providing the foundations by filling in these gaps and unravelling the misconceptions, new learning may literally be impossible for them to understand, increasing frustration and feelings of failure. AI can help identify those gaps, scaffold learning and build understanding.
AI is by no means a replacement for teachers or teaching assistants, it is purely additional support. Coupled with approaches that promote engagement with learning, AI can enable these disadvantaged young people to access an education previously denied them.
According to the DfE, ‘All children are entitled to receive a world-class education that allows them to reach their potential and live a fulfilled life, regardless of their background.’ AI can help support the most disadvantaged young people towards gaining the education they deserve, and creating a pathway towards educational and social equity.
Business
How collaborative robots can support productivity for 2024

By Stacey Moser, Chief Commercial Officer, Universal Robots
The past year has been no easy feat for businesses, with an unpredictable global economy hitting manufacturers particularly hard. As we soon move into 2024, there are no signs of this relenting, with the sector recording it’s worst performance in the UK since the 1980s. Simultaneously, the manufacturing industry is battling 74,000 unfilled vacancies, creating a £6.5 billion economic shortfall which is hindering the ability of companies to fight back.
Increasing productivity will be a key focus in the pursuit of profitability for 2024. Accelerating adoption of automation within the manufacturing process offers a solution, with one study suggesting that automation has the potential to contribute $15 trillion USD to the global economy by 2030. However, many businesses at this time do not have the funds or resources to adopt large, expensive industrial robots.
Enter collaborative robots (cobots), lightweight robotic arms that can automate repetitive tasks usually requiring the skills and manpower of human workers. Compatible with traditional industrial robots commonly employed by manufacturers, cobots work alongside humans to offer a wide range of benefits, all underpinned by the ability to improve productivity, as well as assure staff safety, and improve worker wellbeing.
Reimagining production lines for 2024
Due to the nature of the industry, manufacturing workers often face so-called dull, dirty, dangerous jobs. For example, palletisation – the process of stacking, loading and securing goods onto pallets – has traditionally been a manual operation that requires staff to perform strenuous tasks repetitively. Long term, this can cause issues for employees such as musculoskeletal damage, as well as mental fatigue. As to the factory output, goods of inconsistent quality subsequently become more commonplace.
Working alongside humans, cobots can take on these undesirable tasks, preventing staff injuries while improving manufacturing quality as human error is reduced. With minimal human effort necessary to operate cobots, these autonomous colleagues can help mitigate the labour shortage issue that is currently leaving many manufacturers vulnerable. Cobots enable workers to be more productive as they focus on more valuable tasks that require more cognition, dexterity and reason, in turn unlocking further business value.
As employees take on more rewarding work, job satisfaction naturally improves. In a sector that faces labour gaps, this can go far in improving employee retention. Prospective workers who may have previously been drawn away from the manufacturing industry by the necessity of performing risky and monotonous tasks, may now see a different future ahead, where the prospect of using modern collaborative technologies excites and enthuses. Reducing turnover of employees and creating a stable pipeline of future talent also ensures manufacturers can not only maintain, but increase productivity, as another year of global economic volatility approaches.
Reducing gaps between competitors
With the current cost of living crisis projected to persist through 2024, consumer behaviour is becoming increasingly unpredictable. Cobots provide the flexibility to adapt to whatever circumstances manufacturers find themselves in, whether production must be scaled up or down, or an expansion into new markets is required. Those that cannot swiftly adapt, risk being left behind. Automation therefore becomes a necessity for manufacturers if they want to keep up with the competition.
Not only can cobots provide flexibility in scaling the size of production, but from palletisation and machine tending, to quality inspection, cobots can perform any task without a need for rest. If necessary, cobots can be programmed and redeployed to perform multiple tasks and can switch between these with ease. By eliminating potential pain points between different functions, cobots can slash factory downtime, improving productivity across the board.
With less human labour required to man machinery, the possibility of longer machine tending shifts is unlocked. This could mean the development of 24/7 factories, with downtime kept to a bare minimum. Manufacturers could see goods produced and out of the door faster than ever before, a colossal productivity boost to consider as 2024 approaches.
Challenging current misconceptions
Despite cobots being widely available to manufacturers of all sizes, these modern automation solutions have yet to be considered a possibility by many. Research shows that the biggest hurdles to implementing automation are capital cost and lack of internal knowledge and experience with the technology. However, on the cost issue, it’s been proven that cobots can achieve return on investment (ROI) in as little as 12 months.
A common myth is that cobots are difficult to implement or use. However, the technology is designed specifically with an approachable user interface, meaning employees do not require specialist expertise to operate or interact. Meanwhile, software designed for use alongside cobots, such as Universal Robots’ PolyScope, means that programming cobots to complete tasks becomes intuitive. Modern cobots – such as the UR20 – are lightweight and have a small footprint, eliminating common issues around a lack of factory floor space too.
Finally, the advent of AI and automation tools such as Chat-GPT has fuelled fears that robots may soon takeover human jobs. Manufacturers who fear backlash may be hesitant to begin deploying and reaping the benefits of collaborative automation. It is important to remember that cobots are designed to work alongside humans, not replace them. Using cobots on a factory floor for example, allows employees to work with robots, not like them. As manufacturers look to improve business resilience in 2024, welcoming cobots into the workforce can also facilitate massive productivity gains.
Forecasting for the year ahead
As manufacturers carefully consider budgets for 2024, automation and cobot technologies should be at the very top of the list. Deployment of these technologies is now essential for increasing productivity, quality and efficiency. Organisations that hesitate to invest today risk losing hard-gained momentum to more innovative competitors.
The future of manufacturing is here. And belongs to those who will innovate and automate.

Embedded Finance: The Opportunity Ahead

Hype, Hysteria & Hope: AI’s Evolutionary Journey and What it Means for Financial Services

Using AI to support positive outcomes in alternative provision

The Sustainability Carrot Could be More Powerful Than the Stick!

Hybrid cloud adoption: why vendors are making the switch in 2022 and why you should too
