Artificial intelligence is disrupting the financial market
It is hard to predict the extent to which the financial sector will change in the long term as a result of machine learning and other technology. The fact is that the artificial intelligence (AI) revolution is already in full swing and having an impact on how banks and lenders make investment and financing decisions.
People are extremely bad at predicting the future. Do you remember the film “Back to the Future”? Why aren’t we already flying around with a DeLorean powered by fusion reactors?
That is why discussing AI and its impact on the future of finance is quite an undertaking. To be honest, coming up with a truly accurate idea of how AI will change the way banks, portfolio managers and insurance companies work is impossible. However, what we can say right now is that the AI revolution is already underway – accompanied by major investment. There has been huge growth in using alternative intelligence in both the fintech and traditional banking markets. This will reach approximately USD 27 billion (average annual growth of 23%) by 2026 and USD 64 billion (up 33%) by 2030.
Data-supported analyses for better decisions
AI is revolutionising the way in which banks and lenders make investment and financing decisions. It is helping them to more accurately assess the credit history of borrowers. One of these companies is OakNorth. Its founders successfully tinkered with computing power and models in the area of machine learning, making them efficient to use. As a result, they were able to develop an AI-based credit platform in just seven years. The main objective of the company is to serve SMEs with a credit requirement of between USD 1 million and USD 30 million.
This is how companies such as OakNorth are hitting the spot: on the one hand, this financing segment is largely undersupplied by large banks and other large credit institutions. Owing to regulatory requirements and cuts in staff and funding, such lending is often too expensive. On the other hand, the scale of these loans is too large for private lenders. OakNorth filled the gap by combining technology and machine learning algorithms. They are currently enabling financial institutions to make quicker and better decisions across the entire credit life cycle. Thanks to its data-driven approach, the company provides support for credit analysis and monitoring. With machine learning, the gathering of extensive data sets and lifelong credit history, it is now in a position to model a forward-looking view of the borrower’s financial situation. In contrast to peer-to-peer lenders, OakNorth assumes the balance sheet risk itself.
The company’s profit results from the interest margin between the rates when the loan is taken out and those to be paid by the borrower. Its success is there for all to see: In just seven years, OakNorth has granted business loans to the value of more than USD 9 billion, with only an extremely small amount being defaulted on.
AI in the case of hedge funds…
Another area in which AI is playing an important role are hedge funds. They are using AI on a large scale in their trading strategies for taking arbitrage opportunities on the markets. For years now, Two Sigma, Renaissance and other giants have been developing quantitative models that use billions of gigabits of data in order to identify and predict arbitrage transactions, changes in market sentiment, turbulence in asset classes, etc. Such models mostly concern statistical results. They attempt to assess the probability of a deal succeeding by using a broad range of signals that are in turn applied to various markets and periods. Over the years, quantitative hedge fund houses have also begun to integrate alternative data sources – i.e. sources that are not connected to market prices in any way – into their models. These alternative data sources range from assessments of the impact of tweets by persons involved and the use of key words by CEOs during discussions with analysts in order to predict the performance of their company, to analysis of satellite images of large raw material quarries, ports and warehouses in order to help assess movements of goods.
… and in private markets
One new trend is that databases and AI are penetrating private markets. Even the most successful venture capital funds are now opting to use new technology so that they can better predict which companies from the fund are sure to be outliers. It is no longer enough to rely on internal networks as your primary sources for the deal flow. More recent analysis systems are now using “early detectors” with machine learning in order to gather information about companies by “crawling” their online profiles. That makes it possible to gather data from alternative sources – such as the company’s website, social media platforms, product libraries and news sources – that can supply “signals” of growth based on transaction data.
More recent analysis systems are now using “early detectors” with machine learning in order to gather information about companies by “crawling” their online profiles.
The objective is to predict where the next “unicorn” will come from. This includes dealing more with individuals than with companies. For example, early-stage funds must be the first to find out about talented individuals who are leaving their job at a company or leaving an existing unicorn to set up their own company. Companies such as Specter do just that and, thanks to a multitude of sources and their own algorithms, are able to make their users aware of companies and talented individuals with high growth potential and enable them to seize opportunities first.
Forcing more efficiency
To summarise, it can be said that although it is impossible to tell how much AI will penetrate the financial sector, it is already clear that the sector is going through far-reaching changes.
There is a very good chance that there will be no going back to the “good old ways of doing things”. Too much competition and time pressure to take arbitrage opportunities are forcing public and now also private markets to become more and more efficient. This trend was enabled by technology, with all the positive and negative implications for the decisions associated with it.