Forbes has confirmed that there have been over 154,000 artificial intelligence (AI) patents filed worldwide since 2010. And it’s no wonder there’s a race to deliver new AI-based technologies – CFOs and Financial Directors can expect to see a 22% cost reduction in operating expenses thanks to certain AI applications.
As a result of this activity, there have been many detailed discussions on AI, especially around how it has created a buzz in financial services. AI, as we all would have heard (may be hearing this more frequently, soon), is a transformative technology that is (or will) constantly change the way we perform tasks.
The numbers and the discussion must make sense in terms of application though. So far, we are aware that AI can help financial services in catching sight of financial patterns, eliminating human error from transactions and enabling forecasting to be accurately based on data. These lead to operating and cost efficiencies. The reality is though, that adoption of AI in financial services is still in its early stages.
The time is right
A Microsoft report – Accelerating competitive advantage with AI – highlights that companies risk falling behind foreign rivals unless they use more AI. The report states that organisations currently using AI are outperforming those that are not by 11.5%, up from 5% a year ago. This could be a welcome boost to UK plc at a time of unprecedented economic and political uncertainty.
The report states: “The need for organisations across sectors to move beyond executing small-scale AI projects and become truly AI-enabled is pressing – and one that will play an increasingly vital role in their ability to gain a competitive advantage in the future.”
The time is right to move from experimentation to implementation.
While implementing AI projects, FDs and CFOs must consider getting the right skills and capabilities in the team. Top tips from Michael Wignall, Azure Business Lead, Microsoft UK for moving from AI experimentation to implementation are:
- Do not view AI implementation as just an IT project. Instead, treat it as a business change programme of which technology is a key element.
- Take the time to get people onboard, articulating the reason for the change to stakeholders and the benefits they can expect. No-one should feel they are having AI ‘done to them’.
- Identify and scope a business problem and then plan how AI can help solve it. Try not to kick off with a new business problem either; start with one you already know and understand.
- Do not expect AI to scale organically. You need to build an organisation-wide strategy. Businesses that are purposeful and think about scale are better than those who do a bunch of projects and hope it will just accrue all together and lead to scale as a result.
There are a few uses for AI that have come to light with substantiated actionable and practical use cases, as we outline here.
Accurate forecasts predictions are crucial to both the speed and protection of many businesses. AI can be applied to detect patterns latent in structured and unstructured data to produce breakthroughs that can improve the precision of decisions for finance professionals. Traditional analytic tools are being steadily replaced by machine learning algorithms which help analyse data, predict risks and identify opportunities and predicting customer behaviour to maximise an organisation’s resource allocation towards customer that might deliver the maximum ROI over their lifetimes.
Technology companies like DataRobot help financial teams quickly build accurate predictive models that enhance decision making.
Tata Consultancy Services has even come up with a framework that suggests using feeds from social network sites, blogs, messages from boards, research reports and market feed to create management dashboards.
Predictive analytics would require ensuring that financial department-wide data policies are aligned towards making the data easily accessible, as well as establishing a pipeline to continue a streamlined data collection process.
Detecting fraud and money laundering
The wonderful thing about AI is that it can detect and anticipate patterns in human behaviour. AI makes it possible to evaluate and decode data and spot shady financial actions. This presents with the possibility of moderating forged claims and transactions.
According to IDC $5.6B expected to be spent on AI-enabled solutions by 2022, including automated threat intelligence and prevention systems, and fraud analysis systems. A recent analysis by IHS Markit shows the actual business value of AI is expected to reach $300B by 2030.
A study by Association of Certified Fraud Examiners (ACFE), in collaboration with analytics leader SAS, has predicted that by 2021, nearly three-quarters of organizations (72%) are projected to use automated monitoring, exception reporting and anomaly detection. Similarly, about half of organizations anticipate employing predictive analytics/modelling (52%; up from 30%) and data visualization (47%; currently 35%).
With predictive analytics, organisations can quickly detect fraud and rate transactions on the basis of a wide range of data. AI platforms go through huge databases looking for abnormalities and communicating possible future steps for prevention.
Indian private sector insurer ICICI Lombard is using AL and ML to detect frauds in segments such as motor and health insurance. At the insurer, AI was being initially used to sanction straight forward cases like maternity and cataract, it is now also being used for more complex cases like dengue, where AI is giving initial approval.
AI’s power to handle huge data quickly and meticulously is likely to shift regulatory compliance. AI can be used to assist organisations decipher compliance requirements and take appropriate actions. Persisted application of AI in regulatory compliance can do away with the need for human intervention and thus, human error.
AI’s ability to interact and learn provides it to suit itself as per the inputs and carry out mandates more efficiently and often automatically.
The Demystifying Artificial Intelligence in Risk and Compliance guide from IBM provides several case studies from banks and capital markets firms where AI has been applied in regulation.
AI’s applicability to replicate a human to the closest possibility can be leveraged by FDs towards reporting. AI platforms like Quill helps in financial reporting with natural language generation (NLG) by converting data into plain English to make the financial professionals understand the insights in simple words, just like a human data analyst would do.
AI can improve modern day accounting by helping FDs with financial reporting with regards the processes that involves data. AI extracts financial information from any source, including feeds, uploads, and paper and addresses all the fundamental automation challenges of source data variability, non-standard taxonomies, integration with existing systems, speed of implementation, and scalability.
In 2016 Abe AI, a virtual financial assistant that integrates with Google Home, SMS, Facebook, Amazon Alexa, web and mobile to provide corporate customers with more convenient banking, released its smart financial chatbot for Slack. The app helps users with budgeting, savings goals and expense tracking.
It is not easy to renounce the significance of AI and its applications for CFOs and Financial Directors. AI in a wider sense has myriad uses. However, for AI applications to be beneficial to the financial teams, organisations need to link precise AI techniques to corresponding areas. This approach will deliver the best possible results and create actionable insights.