The global Machine Learning (ML) market is estimated to grow by a CAGR of 43% by 2029, according to a…

The global Machine Learning (ML) market is estimated to grow by a CAGR of 43% by 2029, according to a new report from ResearchAndMarkets.

The major drivers for the ML market are identified as the proliferation in data generation, technological advancements in ML and the increased adoption of connected devices and their data driven application. Enterprises are now awash in data related to their customers, prospects, internal business processes, suppliers, partners and competitors.

The ResearchAndMarkets report found that, often, businesses can’t control this flood of data and convert it to actionable information for growing revenue, increasing profitability and efficient business operation. Organisations of all disciplines across the globe suffer a serious problem of managing data in the form of data retention, understanding dark data, data integration for proper analytics, data access and others.

Data volumes set to rise exponentially

Machine learning offers a promising solution to gain economic benefits from the increase in data with the help of predictive analysis and reducing fraud. The volume of data collected by businesses worldwide is estimated to double every year with a lack of understanding of data now cited as a primary reason that overruns project costs to the tune of 20-35% of operating revenues.

Big data capabilities can assist in providing constant changing customers preferences helping companies to improve customer satisfaction, enjoy faster decision making and develop strategies for launching new products while exploring new markets.

Machine Learning in BFSI remains indispensable

The global Machine Learning market has been segmented on the basis of verticals, deployment modes, organisation size and service. The vertical segment is further sub-segmented into banking, financial services, and insurance (BFSI), retail, telecommunication, healthcare and life sciences, manufacturing, government and defence, energy and utilities and other verticals. The BFSI segment leads the verticals in terms of revenue in the global ML market with around 21.9% market share in 2020.

BSFI is primarily driven by a growing demand for ML to automate the process of loan approval, for fraud prevention, risk management, investment predictions, marketing and other strategies. Prominent banks across the globe including JPMorgan Chase, Wells Fargo, Bank of America, Citibank, U.S. Bank and others have adopted ML to realise the potential benefits of data driven decision making.

Machine Learning in healthcare promises significant opportunities

The healthcare and life science vertical is anticipated to grow at the highest rate over the forecast period (2021-2029) growing at a CAGR of 44.3% over the forecast period. This high growth rate is attributed to the fact that ML solutions offer wide potential across the healthcare industry. This include patient data & risk analysis, in-patient care and hospital management, medical imaging and diagnosis, drug discovery, life style monitoring and management, medical diagnosis and imaging, precision medicine and more.

Additionally, key companies are providing various ML systems across healthcare; these include Google Deep Mind Health, IBM Watson and others. Moreover, increasing healthcare expenditure also leverages huge adoption opportunities for ML. According to the Institute of Health Metrics and Evaluations, global healthcare expenditure is expected to reach $18.28 trillion globally by 2040.

Learn more about emerging trends across the tech panorama in the latest issue of Interface

Gurpreet Purewal, Associate Vice President, Business Development, iResearch Services, explores how organisations can overcome the challenges presented by AI in 2021.

2020 has been a year of tumultuous change and 2021 isn’t set to slow down. Technology has been the saving grace of the waves of turbulence this year, and next year as the use of technology continues to boom, we will see new systems and processes emerge and others join forces to make a bigger impact. From assistive technology to biometrics, ‘agritech’ and the rise in self-driving vehicles, tech acceleration will be here to stay, with COVID-19 seemingly just the catalyst for what’s to come. Of course, the increased use of technology will also bring its challenges, from cybersecurity and white-collar crime to the need to instil trust in not just those investing in the technology, but those using it, and artificial intelligence (AI) will be at the heart of this. 

1. Instilling a longer-term vision 

New AI and automation innovations have led to additional challenges such as big data requirements for the value of these new technologies to be effectively shown. For future technology to learn from the challenges already faced, a comprehensive technology backbone needs to be built and businesses need to take stock and begin rolling out priority technologies that can be continuously deployed and developed. 

Furthermore, organisations must have a longer-term vision of implementation rather than the need for immediacy and short-term gains. Ultimately, these technologies aim to create more intelligence in the business to better serve their customers. As a result, new groups of business stakeholders will be created to implement change, including technologists, business strategists, product specialists and others to cohesively work through these challenges, but these groups will need to be carefully managed to ensure a consistent and coherent approach and long-term vision is achieved. 

2. Overcoming the data challenge

AI and automation continue to be at the forefront of business strategy. The biggest challenge, however, is that automation is still in its infancy, in the form of bots, which have limited capabilities without being layered with AI and machine learning. For these to work cohesively, businesses need huge pools of data. AI can only begin to understand trends and nuances by having this data to begin with, which is a real challenge. Only some of the largest organisations with huge data sets have been able to reap the rewards, so other smaller businesses will need to watch closely and learn from the bigger players in order to overcome the data challenge. 

3. Controlling compliance and governance

One of the critical challenges of increased AI adoption is technology governance. Businesses are acutely aware that these issues must be addressed but orchestrating such change can lead to huge costs, which can spiral out of control. For example, cloud governance should be high on the agenda; the cloud offers new architecture and platforms for business agility and innovation, but who has ownership once cloud infrastructures are implemented? What is added and what isn’t? 

AI and automation can make a huge difference to compliance, data quality and security. The rules of the compliance game are always changing, and technology should enable companies not just to comply with ever-evolving regulatory requirements, but to leverage their data and analytics across the business to show breadth and depth of insight and knowledge of the workings of their business, inside and out. 

In the past, companies struggled to get access and oversight over the right data across their business to comply with the vast quantities of MI needed for regulatory reporting. Now they are expected to not only collate the correct data but to be able to analyse it efficiently and effectively for regulatory reporting purposes and strategic business planning. There are no longer the time-honoured excuses of not having enough information, or data gaps from reliance on third parties, for example, so organisations need to ensure they are adhering to regulatory requirements in 2021.

4. Eliminating bias

AI governance is business-critical, not just for regulatory compliance and cybersecurity, but also in diversity and equity. There are fears that AI programming will lead to natural bias based on the type of programmer and the current datasets available and used. For example, most computer scientists are predominantly male and Caucasian, which can lead to conscious/unconscious bias, and datasets can be unrepresentative leading to discriminatory feedback loops.

Gender bias in AI programming has been a hot topic for some years and has come to the fore in 2020 again within wider conversations on diversity. By only having narrow representation within AI programmers, it will lead to their own bias being programmed into systems, which will have huge implications on how AI interprets data, not just now but far into the future. As a result, new roles will emerge to try and prevent these biases and build a more equitable future, alongside new regulations being driven by companies and specialist technology firms.

5. Balancing humans with AI

As AI and automation come into play, workforces fear employee levels will diminish, as roles become redundant. There is also inherent suspicion of AI among consumers and certain business sectors. But this fear is over-estimated, and, according to leading academics and business leaders, unfounded. While technology can take away specific jobs, it also creates them. In responding to change and uncertainty, technology can be a force for good and source of considerable opportunity, leading to, in the longer-term, more jobs for humans with specialist skillsets. 

Automation is an example of helping people to do their jobs better, speeding up business processes and taking care of the time-intensive, repetitive tasks that could be completed far quicker by using technology. There remain just as many tasks within the workforce and the wider economy that cannot be automated, where a human being is required.

Businesses need to review and put initiatives in place to upskill and augment workforces. Reflecting this, a survey on the future of work found that 67% of businesses plan to invest in robotic process automation, 68% in machine learning, and 80% investing in perhaps more mainstream business process management software. There is clearly an appetite to invest strongly in this technology, so organisations must work hard to achieve harmony between humans and technology to make the investment successful.

6. Putting customers first

There is growing recognition of the difference AI can make in providing better service and creating more meaningful interactions with customers. Another recent report examining empathy in AI saw 68% of survey respondents declare they trust a human more than AI to approve bank loans. Furthermore, 69% felt they were more likely to tell the truth to a human than AI, yet 48% of those surveyed see the potential for improved customer service and interactions with the use of AI technologies.

2020 has taught us about uncertainty and risk as a catalyst for digital disruption, technological innovation and more human interactions with colleagues and clients, despite face-to-face interaction no longer being an option. 2021 will see continued development across businesses to address the changing world of work and the evolving needs of customers and stakeholders in fast-moving, transitional markets. The firms that look forward, think fast and embrace agility of both technology and strategy, anticipating further challenges and opportunities through better take-up of technology, will reap the benefits.

With virtually all companies looking at AI, what are some of the key risks they need to consider before implementation?

Today virtually all companies are forced to innovate and many are excited about AI. Yet since implementation cuts across organisational boundaries, shifting to an AI-driven strategy requires new thinking about managing risks, both internally and externally. This blog will cover “the seven sins of enterprise AI strategies”, which are governance issues at the board and executive levels that block companies from moving ahead with AI. by By Jeremy Barnes, Element AI

1- Disowning the AI strategy

This is probably the most important sin. In this case, a CEO and board will say that AI is a priority, but delegate it to a different department or an innovation lab. However, success is not based on whether or not a company uses an innovation lab—it’s whether they are truly invested in it. The bottom line is that the CEO and board need to actively lead an AI strategy.

2- Ignoring the unknowns

This happens when companies say they believe in AI, but don’t reach a level of proficiency where it’s possible to identify, characterise and model the threats that emerge with new advances. Even if it is decided not to go all-in on AI innovation, it’s still important that there is a hypothesis for how to address AI within a company and an early warning system so the decision can be re-evaluated early enough to act.  Being a fast follower requires as much organizational preparation and lead time as leadership.

3- Not enabling the culture

The ability to implement AI is about an experimentation mindset. That and an openness to failure need to be adopted across the company. Organisations need to keep in mind that AI doesn’t respect organisational boundaries. Most companies want high-impact, low-risk solutions that could simply lead to optimising, rather than advancing new value streams. It is hard to accept increased risk in exchange for impact but it will come as part of the continuous cultural enablement of an experimental mindset.

4- Starting with the solution

This is the most common sin. It’s important to be able to understand the specific problems you’re trying to solve, because AI is unlikely to be a solution for all of them, and especially not blindly implementing a horizontal AI platform. Have the conversation at board level to ensure that an overarching AI strategy, and not simply quick-fix solutions, is the priority.

5- Lose risk, keep reward

As mentioned in the third sin, it is natural for companies to want to implement AI without any risk. But there is no reward without risk. A vendor motivated to decrease risk will also decrease innovation and ultimately impact by making successes small and failures non-existent. AI creates differentiation only for companies that are willing to learn from both their successes and their failures. A company that doesn’t effectively balance risk in AI will ultimately increase its risk of disruption.

6- Vintage accounting

Attempting to fit AI into traditional financial governance structures causes problems. It doesn’t fit nicely into budget categories and it’s hard to value the output. The link between what you put in and what you get out can be less tangible or predictable, which often makes it harder to square with existing plans or structures. Model the rate of return on AI activities and all data-related activities. This demands that these activities affect profit (not just loss) and assets (not just liabilities).

7- Treating data as a commodity

The final sin concerns data and its treatment as a commodity. Data is fundamental to AI. If data is poorly handled, it can lead to negative impacts on decision-making. Data should be treated as an asset. The stronger, deeper and more accurate the dataset, the better models that you can train and more intelligent insights you can generate. But, at the same time, when personally identifiable information is stored about customers, it can be stolen, risking heavy penalties in some jurisdictions. You need to build towards data from a use case rather than invest blindly in data centralisation projects. So, now you know what not to do. Here are some of the simple things that you can do to move ahead. First, talk to your board about how long it will take to become an AI innovator, modelling it out, rather than simply discussing it conceptually.

Second, prepare for change and put in place monitoring. AI shifts all the time, so you’ll want to regularly check in to adjust and pivot your strategy. It’s important to develop a basic skill set so you can redo planning exercises with your board. Third, model out risks in both action and inaction. But don’t model them in a traditional approach, which is to push risk down to different business units and then compensate those units for reducing risk rather than managing trade-offs. Instead, view those trade-offs in terms of risks and rewards, and start to think about how you are accounting for the assets and liabilities of AI. Ultimately, you want to start to model what is the actual rate of return for all these activities that you are doing. Then benchmark it against what you see in other companies from across the industry, and that will give you a good picture of the current situation and where to go.

NexBotix, the Robot-Process-Automation (RPA) service, has officially launched in the UK. With a managed dashboard solution applied to specific business objectives, it means that only the right processes are automated and ROI can be delivered in as little as 30 days.

NexBotix delivers a low-cost solution for businesses across finance and accounting, HR, IT, governance and compliance departments, across banking, financial services, insurance, automotive, logistics, legal, retail and local government. Unlike anything else currently in the market, the platform can be deployed into existing IT infrastructure in just 14 days.

NexBotix uses today’s leading technology from major vendors such as Microsoft, Google, IBM Watson, Automation Anywhere, NICE, UiPath and Abbyy, alongside its own NexBots. The platform provides businesses with the ability to scale up and down their operations according to demand and assist teams in focusing on more high-value tasks, all the while, driving down cost. The key to the multi-vendor approach, is the NexAnalytics capability that helps companies gain complete control of their digital workforce and ensure that the Business Case ROI is delivered as specified.

Chris Porter, CEO of NexBotix, says: “The ‘plug, play, and managed’ element of our technology means that there’s minimal disruption to existing operations, and with no-code to manage it doesn’t require users to be tech-savvy to operate it. With some of the more established players in the market, there’s typically a three month consultation period before any integration can begin, so is it any wonder that enterprises are becoming disillusioned with the actual impact automation can have? We’re so confident in our technology and team that we offer customers a guarantee of receiving ROI within three-to-nine months; though in many cases we’ve seen this happen within as little as a few weeks.”

“With NexBotix, it’s less about removing the human element, but more about working alongside process automation to arrive at the best possible outcome; both in terms of efficiency and profitability. Where most businesses fail with AI implementation is that they lack the foundations intrinsic to its success as a model. Where NexBotix differs is that we put a specific business situation first, and build around that.”

The platform is managed by a team of experts within NexBotix, so it removes the need for any company to have a dedicated technical resource and the service can deliver quantifiable benefits 30 days from implementation. In one case, NexBotix helped a customer service organisation with 3,000 employees achieve an ROI of 802% and payback within four weeks, for its sales department.

Nexbotix has been spun out from Camwood Ltd which has over 20 years of experience and a proven portfolio of products and services across intelligent automation. Most notably, it sold AppDNA to Citrix in 2011 for $91.3m.

Everyone wants to implement Artificial Intelligence (AI) and Business Intelligence (BI) solutions. AI alone is anticipated to generate $15.7 trillion…

Everyone wants to implement Artificial Intelligence (AI) and Business Intelligence (BI) solutions. AI alone is anticipated to generate $15.7 trillion in GDP by globally 2030, and as this market grows, AI and BI will shift from industry buzzwords, to key market differentiators, before eventually becoming the new normal in the corporate landscape.

Yet bringing AI and BI on board is a big leap if it’s your first major data project. Stibo Systems’ Claus Jensen, Head of Emerging Technology, comments on the role of MDM as a vital foundation to implement emerging data technology.

Most CEOs don’t trust their own data.*

Let that sink in for a moment.

Almost every business is looking to data solutions to fuel the next phase of growth and innovation. AI and BI are firmly on the agenda, yet a report by Forbes Insights and KPMG found 84% of CEOs are concerned with the quality of the data they’re basing their decisions on.

That’s a significant disconnect. Businesses at board level want to implement ‘next generation’ data projects, but don’t trust the data that will be fed into them. For CDOs and other data leads, this presents a difficult situation. They need to meet demand for cutting-edge data projects, knowing that there is a certain level of mistrust in the data at their disposal.

For many CDOs, that mistrust isn’t limited to the CEO. Think about the data you are currently processing: how confident are you that it’s being accurately sourced, entered, saved, stored, copied and presented? How well do you know that data journey once it leaves your sphere of control? Are you certain that a single source of truth is being maintained?

The data gold rush

It may only be major data breaches that make the headlines, but in the global gold rush for data, too many businesses fail to accurately extract, store and interpret data.

Mistakes are made at every stage in the process – in fact, so bad are we at processing data, a report by Royal Mail Data Services claims that around 6% of annual revenue is lost through poor quality data.

It’s equally bleak in the US, where Gartner’s Data Quality Market Survey puts the average cost to US business at $15 million per year.

Despite this, we’re rapidly moving the conversation from data capture to artificial intelligence (AI), business intelligence (BI) and connected devices (IoT) – and for good reason.

Putting aside the issue of bad data (we’ll come back to that), businesses now have access to more data than they can handle – according to SAS’ Business Intelligence and Analytics Capabilities Report, 60% of business leaders struggle to convert data into actionable insights, and 91% of companies feel that they are incapable to doing it quickly enough to make useful changes. 

Business Intelligence and Analytics Capabilities Report

In large businesses, where data streams are blended from many sources, machine learning can help data scientists monitor figures to flag outliers, irregularities and noteworthy patterns.

Once flagged, business leaders can use BI to bring those patterns to life, helping pave the way for the most appropriate, and profitable, action.

Stibo Systems’ Head of Emerging Technology, Claus Jensen, believes it’s only a matter of time before we see AI regularly used within business product features – with machine learning automating tasks thanks to effective data interpretation.

Jensen and his team are working at the forefront of data: building master data management solutions in conjunction with AI and BI. “We’re entering into a new era of data analytics,” says Jensen. “Data scientists aren’t going away, but they can do more and more high-level work as certain use cases are solved by AI.” 

One of these use cases is machine learning-based auto classification. “For retailers onboarding thousands and thousands of new products every month, it’s really time consuming for them to have the vendor categorise the product into the vendor taxonomy.

“Machine learning can automate this based on product description and image.”

Running before we can walk

As exciting as this sounds, businesses eager to install new uses for data often face significant challenges: their data isn’t watertight, or it’s siloed, often both.

In a piece penned for the Financial Times, Professor of Economics at Stanford Graduate School of Business, Paul Oyer, wrote: “Smart managers now know that algorithms are as good as the data you train them on.” In other words, AI (and analytics for that matter) can only ever be as good as the date you feed it.

Which brings us back to the question of trust. What needs to happen for CEOs to trust their own data?

While there’s no single answer to this question, a master data management (MDM) solution is a good place to start.

“You can think of MDM as the foundation, a layer, that provides a single source of the truth for data,” explains Jensen. “Analytics and machine learning is only useful if the data you’re working on is accurate. That’s where MDM comes in; it ensures information presented, and actions taken, are based on fact and reality.

“Otherwise, business analytics is just a nice and colourful way to look at bad data, and what’s the point in that?”

To find out more about how MDM can turn data into business value through actionable insights, forming a solid foundation to AI and BI, visit

Welcome to the June issue of Interface Magazine! Read the latest issue now! This month’s cover features Gary Steen, TalkTalk’s…

Welcome to the June issue of Interface Magazine!

Read the latest issue now!

This month’s cover features Gary Steen, TalkTalk’s Managing Director of Technology, Change, and Security, Gary Steen regarding the telco’s commitment to thinking, and acting, differently in a highly competitive marketplace…

TalkTalk is an established telecommunications company that fosters a youthful, pioneering spirit. “I like to think of TalkTalk as a mature start-up,” says Managing Director of Technology, Change and Security, Gary Steen. “We are mature in terms of being in the FTSE 250, with over four million customers, relying on our services every day through our essential, critical national infrastructure. But that said, I definitely think we start our day thinking as a start-up would. What can we do differently? How do we beat the competition? How do we attract great talent? We’ve got to come at this in a different way if we are going to succeed in the marketplace. We are mature, but we think like a start-up.”

Elsewhere we speak to Natalia Graves, VP Head of Procurement at Veeam Software who reveals the secrets to a successful procurement transformation. Graves was tasked with looking at the automating, simplifying, and accelerating of Veeam’s procurement and travel processes and systems around them, including evaluating and rolling out a company-wide source-to-pay platform. “It has been an incredible journey,” she tells us from her office in Boston, Massachusetts. We also feature exclusive interviews with PTI Consulting and cloud specialists CSI.

Plus, we reveal 5 of the biggest AI companies in fintech and list the best events and conferences around.

Enjoy the issue!

Kevin Davies