This article is based on the keynote speech presented by Jean-Phillipe Desbiolles at the CFTE’s Virtual Campus event, namely – “AI in a post-COVID world: Rupture or Acceleration?”. Jean-Phillipe is IBM’s Global Vice-President for Data, Cognitive & AI – Financial Services.
In light of the current unprecedented times, this article aims to assess the trajectory of AI adoption in the financial services industry, alongside shedding light regarding the process of which AI implementation is carried out within businesses seeking to digitally transform.
What is AI?
To recap, let’s first look at the concept of Artificial Intelligence. It is difficult to pinpoint exactly what the definition of AI is—everyone has their own variation.
According to a Deloitte report, the non-technical definition of AI includes the following.
Artificial Intelligence is a suite of technologies, enabled by adaptive predictive power and exhibiting some degree of autonomous learning, that dramatically advance our ability to:
- Recognise patterns
- Anticipate future events
- Create good rules
- Make good decisions
- Communicate with other people
Alternatively, a technical definition that we can offer is that artificial intelligence most resembles a Neural Network is system software that works similar to the tasks performed by neurons of a human brain. In other words, Artificial Intelligence refers to the ability of a computer program or a machine to think and learn.
To illustrate, we can look at the AlphaZero study that showcased an AI algorithm that was able to beat the strongest available chess engine (Stockfish) after only 4 hours of self-play.
AI in Finance: Rupture or Acceleration?
Pre-COVID 2020 predictions
In 2019, a report by IBM: Roadblock to Scale identified that 2020 would be the year that companies scale to scale their AI and data capacities to further transform their business processes. Namely, the industry has matured to a certain point of growth, bypassing the former stages of experimentation and initial adoption.
For this to happen, there are 3 factors that need to be achieved:
- Organisations have to redesign and build the necessary operating models to scale their AI and data capacities.
- Organisations have to transform and adopt the appropriate data foundations by redesigning robust DataOps and a data platform.
- AI needs to be trusted, transparent and explicable.
The report states that 75% of EU businesses are already implementing AI and 47% of them plan to have AI deployed across the business. However, organisations have stumbled upon a range of barriers in their AI adoption, namely, a lack of skills (37%), data complexities/silos (31%), a lack of the right technical tools (26%).
Further, trust has also become a critical issue. For a seamless integration of AI with business processes, 78% of respondents state their need to be able to trust that their AI outputs are reliable and accurate whereas 83% require an explanation behind every output produced by AI.
On an AI and data perspective, this global health crisis impacts banks and insurers on 3 levels:
- Client experience
Of the late, we have been bombarded by advice from health experts regarding ‘the new normal’. Namely that, prior to a vaccine being completed, individuals are advised to adapt to the new normal of increased hygiene, social distancing and wearing face masks in public.
On the other hand, however, for AI in a post COVID world, the new normal will largely consist of acceleration.
For example, for client experience post-COVID, consumers would expect an increased end-to-end digital user journey, which leveraged AI and data to improve seamlessness, simplicity, personalisation and security.
In terms of workflow, COVID19 has highlighted the need for operational continuity and resiliency. Thus, this has made AI and data redesign of intelligent workflows the latest business process re-engineering. Some examples include: robotic process automation (RPA) and learning process automation (LPA).
Post-COVID culture on the other hand, will witness an increase in remote working and/or working from home. Following this, there will be a demand for remote work guidance and governance, as well as additional tooling and training to cope with the changes in client experience and workflow.
Real-case examples of scaling AI and data capacities include: machine learning models increasing accuracy for OCR, insurance claim and credit scoring by 20-40% and AI chatbots for an enriched CRM.
Thus, COVID has only served to accelerate the adoption and scaling of AI in finance.
Operating model to scale AI and data within an organisation
- Innovation: Identify, assess and prioritise new use-cases
- Industrialisation: Training and development of new AI solutions
- Assets: Identify reusable blocks and facilitate the acceptance of these solutions by users
- Continuous improvement: Accelerate and industrialise the creation of cognitive solutions
- HR: Attract, retain and reskill talents. Enable adoption and change management
Acceleration key points
- Rethink the whole user experience: Be holistic, don’t think in silos and prioritise simplicity
- Integration is the true value: It is only through true integration can there be an effect on business performance
- Place human beings at the centre of everything: Continuous learning for employees is a priority. If organisations’ employees do not reskill themselves accordingly, the digital transformation will not succeed fully.
Looking to future-proof your organisation?
CFTE’s bespoke programmes are crafted with leading experts from the largest global universities, organisations and startups – curated for the industry, by the industry. We specialise in digital transformation, and our signature programmes include AI in Finance and Fintech Foundation. They cover topics of Artificial Intelligence (AI), Machine Learning (ML), data analytics, Robotic Process Automation (RPA) and more. CFTE is also currently working on developing new programmes on Open Banking and Regtech, get in touch to learn more.