The world of artificial intelligence (AI), particularly generative AI, exploded in 2023 – following the hype and interest in OpenAI‘s ChatGPT. But as we foray into 2024, what does the future hold for the space? We turn to the experts to find out.
Off the back of a year dominated by discussion about AI, it is easy to wonder what trends lie in store for us in 2024. With all types of possibilities for all types of industries, it appears as though AI could impact absolutely every aspect of business, finance and beyond.
In 2023, a McKinsey Global Survey covering the ‘State of AI‘ found that 40 per cent of respondents believed their organisations would increase investment in AI overall because of advances in gen AI. Released in August, these findings are likely even more true now. We’ve seen a rapidly increasing number of firms embracing AI technology and exploring how it can be used in every facet of their operations.
“The rise of certain technologies like generative AI in 2023 has heralded a great deal of transformative change that will shape how the payments industry develops in 2024,” explains Tony Craddock, director general of The Payments Association.
To better understand just how much of an impact AI could have throughout the coming year, we hear from a range of experts in the industry to get their takes on the most important upcoming AI trends for 2024.
AI: The here and now
As Iain Armstrong, regulatory affairs practice lead for ComplyAdvantage, the AI-driven fraud and AML risk detection firm, explains, it’s no longer a question of ‘if’ for AI, more ‘when’ and ‘how’: “As we head into 2024, the question is no longer if companies invest in AI, but what kinds of skills their analysts need to ensure that the models they use are effective and that they can justify decisions that they make to auditors.
“Key skillsets such as data preprocessing, model performance monitoring and optimisation, and experience in automated decisioning strategies will be in demand. Staff in existing anti-financial crime roles will benefit massively from gaining a base-level understanding of machine learning and AI. Companies that invest in staff training in this area will reap the dividends.”
Michael Conway, executive partner of data, AI and technology transformation service line leader at IBM UK & Ireland, also expects 2024 to be a year of AI-related progress for firms.
Conway explains: “2023 was the year of experimentation with generative AI, but not many businesses were able to put new AI models into production. In 2024, it will be about ramping up the production of new generative AI use cases where the real commercial value lies.
“This in turn will drive a trend for AI platforms with built-in data and AI governance capabilities that help businesses comply with regulations, trace the lineage of their data and explain how the outputs of their AI models were generated.”
‘A fresh start for many organisations’
Continuing the theme of cautious optimism regarding what 2024 holds in store for AI, Eldar Tuvey, CEO and co-founder of spend management platform Vertice, explains: “Leaving behind a belt-tightening 2022 and 2023, the new year presents a fresh start for many organisations as they enter a period of lean, efficient growth.
“Automation is high on the list for executives, but I expect to see faster adoption of AI software amongst CFOs seeking efficiencies on everything from budgeting and invoicing to analysing contracts at speed. The finance function has often been described as the canary in the coal mine for new technology as teams are eager to unburden themselves from manual, time-consuming tasks and find time for strategic projects.
“The pressure to keep up with new software also presents a budgeting challenge for the modern CFO, who will need to devote more scrutiny and attention on software purchasing in general to keep tight control of spending.
“Legacy products will be under the spotlight as new players emerge, presenting a potential new dawn in lots of SaaS markets with companies less willing to stay loyal to the big software players if they can’t keep up with the AI arms race.”
2024: The year of AI feature rollouts?
Both Dave Tonge, CTO of Moneyhub, and Elliot Colquhoun, VP of Information Security and IT at Airwallex, suggest that most companies are readying themselves
Tonge explains his 2024 prediction: “We will continue to see the rollout of ‘AI features’ into a wide range of products. Some of this will be gimmicky, but some will be useful productivity boosters. There will continue to be hype and fear around AI, but this is a tech innovation that is already solving a wide range of problems (something crypto never managed!).
“There will also be an increase in ‘agent’ based tools. These are AI-powered systems that can solve complex problems by calling multiple tools. Rather than simple input-output chains, these agents will leverage advanced algorithms to orchestrate a variety of data sources and AI models. Agent-based AI systems will not only predict outcomes but also recommend and execute actions, leading to more autonomous and intelligent software.”
‘This year, we can expect AI to advance exponentially’
Colquhoun adds: “In 2024, I think we’ll see even greater adoption and rollout of AI-first features by almost every company. Similar to tech trends of the past, mass uptake will also lead to mass innovation. We’ve already seen widespread interest and adoption of generative AI for customer support, exploring data, writing code and performing increasingly complex tasks.
“This year, we can expect AI to advance exponentially in terms of the complexity and specificity of tasks it can perform. Entrepreneurs will go deeper, finding ways to apply AI to solve problems, particularly for non-technical users, supercharging their output.
“For example, the fintech and financial services industry must comply with strict regulatory and compliance guidelines that often heavily rely on manual human processes across information security, compliance support, know your customer (KYC) onboarding, transaction monitoring and more. Processes that are manual, error-prone, time-consuming, and hard to automate with legacy approaches – such as comparing a change management ticket to the actual changes, or reviewing a description of a business – are natural applications of AI that completely change our approach to solving problems.
“As with all emerging technologies, businesses need to take a pragmatic approach to integrating AI into their products and operations. Although there are new risks to address, there are also compelling new opportunities to reduce risks – new ways to prevent fraud, investigate alerts faster, and keep customers safe. If we can adopt AI thoughtfully, we can not only ensure we’re using AI safely but improve safety in other areas too, such as customer onboarding and fraud prevention.”
Regulation and responsible AI
The importance of regulation and using AI responsibility can also not be overstated. Michelle Moody, MD of management consulting firm Protiviti UK, explains: “Undoubtedly, additional foundational models will be released to help organisations adopt AI quickly and be better trained for specific industries.
“The pace of adoption in 2024 will be aligned with the pace of additional regulation and legislation, as well as the maturity of the organisations’ data estate and the advances in technology.
“I would also expect organisations will be looking at adopting AI safely so that it is transparent and explainable, especially with the first laws being introduced to manage risk.
“Data is another important area that may need to be discussed more. AI models are trained with vast amounts of data where patterns are identified, and models are coded to make decisions using that data. If the data is not fit for purpose, or the coding is biased in making decisions, then the model will not provide quality outputs. So, the old adage of garbage in-garbage out applies here,” continues Moody.
“If organisations do not have strong data governance and accountability for the data from C-level down to every employee, then they may find themselves falling foul of legislations and potential fines coming into effect. Organisations will need to collect data responsibly and ensure that the data is being used ethically and aligned with current privacy laws. Adopting AI without good-quality data to train the model will be difficult.”
