Quant funds are nothing new and Exchange-traded funds (ETFs) that rely solely on artificial intelligence (AI) to make investment decisions are today’s news today. It is most likely that AI and machine learning will continue indefinitely in the future of investment decision-making for asset managers. There is no denying that the potential of AI in asset management is substantial and one must question how will man be able to keep up with the machines in terms of risk monitoring and compliance?
By definition, there is a “black box” aspect to machine learning, in that there is a lack of transparency into why the technology has identified a certain pattern. Over time, this complexity only increases and makes it difficult for compliance and risk teams to understand and manage. This issue is magnified with heavily-regulated Ucits funds. There is no doubt that AI optimises operating models, but it can be hard to understand and explain when a “black box” has delivered the outputs.
We find ourselves in an era of unprecedented advances in the field of artificial intelligence and the impact that is having on fund managers and man-cos cannot go unnoticed. Some of the world’s largest funds now heavily rely on artificial intelligence to make investment decisions, so it makes sense to ask why Ucits fund managers and management companies don’t make more use of artificial intelligence to lighten their Ucits regulatory and compliance load. Oftentimes the answer is that AI outputs are too difficult to comprehend and that’s where XAI comes in.
Explainable AI (XAI) may well be revolutionary for compliance and risk teams. Firms with compliance and risk departments equipped to decipher the interpretability and explainability of AI will have the most powerful, regulatory-compliant models. In the world of data science interpretability requires an understanding of the inner layers of models, in tandem with easily understandable summary form outputs for the lay user. XAI platforms with enhanced data insights at the core and simple effective automated and manual workflows are the future for compliance and risk teams.
For Ucits funds XAI can springboard automation opportunities across compliance monitoring in areas that are currently labour-intensive, time-consuming and error-prone. With XAI gone are the days of compliance or risk officers spending hours reviewing regulatory and compliance risk across Ucits funds. XAI can both mine complex data for risk-relevant facts, and simplify the process of identifying high-risk. Not only that but XAI can further aid the compliance or risk officer in their understanding as to why the risk is high.
XAI is particularly valuable when performing time consuming, repetitive tasks. No more manual compliance hours need to be spent reconciling NAV packs to portfolio holdings to ensure that the Ucits rules aren’t being breached. With XAI the data is mined, modelled, analysed, and then output in a manner that even the most junior of staff can explain and understand. In a world were risk and compliance talent retention is challenging, savvy Ucits managers and management companies are embracing XAI technology to retain their key risk and compliance personnel.
XAI has the ability to track the changes in Ucits regulations, identify information gaps and provide an easily comprehensible and detail oriented alerts which can then be used to perform compliance outreach to fund administrators, fund service providers and investors alike to collect the outstanding information. This truly revolutionises what was once either an automated “black box” process or a tiresome manual workaround.
There are numerous ways that XAI can revolutionise Ucits risk and compliance processes:
- Counterparty risk profiling
XAI can automate the creation and updating of the counterparty risk profiles during post trade compliance and match this against the classification process (i.e. high-, medium- and low-risk). The ranking can then be displayed and explained on screen through a number of visualisation techniques that ensure continued compliance through all dealings with the counterparty.
- Issuer risk profiling
XAI’s ability to “read” vast amounts of data (including unstructured text) and derive meaning can help in producing comprehensive, accurate and auditable risk profiles on companies and issuers in a matter of minutes. This adds huge benefit to compliance teams who are tasked with understanding issuer risks and improves their ability to draw accurate conclusions for a risk-based approach to compliance.
- Compliance investigation
Alert management is challenging for compliance teams. When applied to workflow automation, XAI has the ability to transform the generation of alerts, compliance reports, audit trails and documentation that subsequently get reviewed by board members, investors and regulators alike. Underpinning the alert generation process with XAI can result in easily understandable alerts and XAI can help to leverage previously performed steps in the alert investigation process formulating an easy to follow step by step approach to compliance investigations.
- Regulatory change management
XAI’s ability to detect patterns in a vast amount of text enables it to form an understanding of the currently changing Ucits regulatory environment. Extraction of useful information such as company news, products and issuer activity impacted by regulatory change, keeps fund managers and mancos on the cusp of regulatory changes.
- ESG Monitoring
With the pending review of Ucits regulation it is likely that all funds registered for sales in the EU and their fund managers will be forced to integrate environmental, social and governance (ESG) criteria in their fund management process and to disclose how the respective fund is performing with regard to upcoming EU taxonomy and/or the Sustainable Development Goals (SDGs) of the United Nations. This means that their compliance oversight in the field of ESG must improve greatly too.
At present investment managers within the ESG sector look to gain an investment edge by using AI. Computers can go through vast arrays of ESG data in a sector where information is often presented in unstructured ways. By using AI to scrape for keywords across the worldwide web ESG companies and emerging ESG technologies can be found and the information can be used to inform investment decisions. Combining the results with explainable output techniques, moving from AI into XAI, is a sure way of helping the compliance team also make use of the ESG data that informed investment decisions. Getting down into the details on ESG funds is a challenge indeed, particularly when one does not use data science to look deep into issuers, companies and their products. For this reason, if funds are using AI for investment decision making, taking one step further and using XAI will help the compliance officers understand ESG investment decisions too.
In summary, Ucits breach monitoring is a challenge for many firms, particularly in a landscape of Ucits regulatory change and compliance talent is scarce in the market. There is, however, another way to overcome the challenge and it doesn’t have to be through people. When compliance challenges make use of AI models that exist in firms for investment decision making and these models are counterbalanced with interpretability and explainability, that’s where the magic happens and the world of Ucits investment breach monitoring is revolutionised. Many firms would say that explainable and transparent compliance monitoring systems are difficult to build. They would not be wrong. However, there are technology vendors in market who have done the hard yards and built XAI solutions so the firms don’t have to. The important thing to know is what you are looking for when you go to revolutionise your Ucits risk monitoring through AI. That is, make sure you look for XAI.
Geraldine Gibson-Dautun has over 20 years FinTech experience and in November 2018 won FinTech Champion of the Year at the Women in Finance awards. Educated at Trinity College Dublin and the Sorbonne University, France, Gibson-Dautun founded AQMetrics in 2012. The company has been recognised as one of the top 100 global regtechs for the last two years running and won the HFM European Technology Award for Best Regtech Solution 2018.