Paris Fintech Forum is undoubtedly one of the most sought-after Fintech events in Europe. With 2,700+ attendees from 75+ countries in 2019, it gathers the best Fintech minds to collaborate and discuss the new industry trends. This year, the event was even bigger and better, with Emmanuel Macron himself taking the patronage of the event. With speakers including Frédéric Oudéa, CEO of Société Générale; Nick Stronovsky, the CEO of Revolut; and Hikmet Ersek, CEO of Western Union, the event was an obvious must-attend for CFTE.
AI in Finance Roundtable
We were honoured to organise an exclusive roundtable during the Forum to discuss the pressuring topic on how AI transforms business models in Finance. We invited some of the most prominent leaders on Fintech transformation, including Ronit Ghose, Global Head of Research at HSBC and Christophe Chazot, ex global head of Innovation, to gather their views on the topic. Chaired by CFTE’s Co-Founder, Huy Nguyen Trieu, the round table discussion generated some great points which we will summarise below.
What is AI?
It is hard to define, as it possesses a broad definition which differs from one academic paper to another. To illustrate, Huy highlights an AlphaZero study that showcased an AI algorithm beating the reigning world champion in chess. From this, it is demonstrated that Neural Networks are very relevant to what Artificial Intelligence actually is.
A Neural Network is system software that works similar to the tasks performed by neurons of a human brain. Besides AI, the other technologies associated with it are Deep Learning and Machine Learning.
Recently, we have witnessed the creation of various new business models. For example, AirBnB leverages the sharing economy while Facebook and YouTube are platforms.
3 key factors that resulted in new Business Models:
- Artificial Intelligence
- New analytics (think of how Facebook leverages user data)
- Automation tools
Interestingly, there have also been some advancements in the utilisation of AI and Data analytics (AIDA) combined, resulting in a more complete framework to apprehend new Business Models.
Paradox raised:
During the summit, various CEOS and CFOs advocated strongly on the ways in which their banks/companies leveraged robust Digital Transformation & AI strategies. However, upon closer inspection of their activities, it is difficult to shed clarity on how exactly these organisations utilise AI, if they really do at all.
Historical perspective:
The concepts currently discussed are not necessarily new. The term ‘Artificial Intelligence’ was coined in 1956. Similarly, in the late 1950s, the idea that information might be more valuable than money had already emerged.
Thoughts from the discussion:
There are some nuances between how Supervised Learning vs Unsupervised Learning can disrupt Business Models. Different points of views were raised regarding the robustness of reinforcement learning.
Supervised Learning and Unsupervised Learning are both Machine Learning techniques. In Supervised Learning, the machine is trained using data that is ‘labelled’, whereas Unsupervised Learning is when the machine is left unsupervised. Supervised Learning is a simpler method of training, Unsupervised Learning is undoubtedly a more complex process.
On the other hand, reinforcement learning—in the context of AI—is a marriage between psychology and dynamic programming that trains algorithms using a system of reward and punishment.
Applications of AI in Finance:
- Optical Character Recognition used for cheque processing in the 80s
- Supervised Learning algorithms trained to recognise fraud
- Facial Recognition for onboarding processes
- AI used to improve risk measurement
- Credit ratings faster and more accurate with AI
Thoughts from the discussion: what is preventing a quick implementation of AI tools in Banks?
- The bulk of big companies/banks are very conservative and not agile enough
- The regulatory sphere is complex and heavy for Banks.
- When people talk about AI, they often picture it as a utopia. This prevents them from leveraging it realistically.
Where is AI the most valuable today and what are the areas with the biggest potential for AI?
- In cases of market failure, AI can be highly valuable
- AI can mitigate against anti/adverse selection in insurance
- AI can empower behavior-based lending
- AI and Machine Learning could be the next frontier for ETFs to outperform the market
It was certainly a fruitful discussion, of which yielded various interesting points about AI in finance. We were delighted to be joined by key opinion leaders in the industry and being able to host an event of this magnitude was a milestone for CFTE.
The biggest takeaway from the roundtable is that the benefits of which AI implementation could bring forth to banks and financial institutions are wide-ranging. However, in order to truly progress, the current biggest pain-point of path dependency—also known as resistance to change—has to be mitigated against.
Follow us on LinkedIn, Twitter and Facebook for more updates from the world of Fintech!
Latest posts:
- Best Practices for Data Considerations | Generative AI in Wealth Management
- Main Challenges in Wealth Management | Generative AI in Wealth Management
- Future-Proofing with AI: A Glimpse into the Future of Cybersecurity
- Current Applications of Generative AI in Cybersecurity
- CFTE Launches the Generative AI for Compliance in Financial Services Programme