What eflow saw at XLoD London - The challenges, insights and questions still to be answered

Last week, the eflow team was delighted to be part of XLoD Global conference in London, an event that brings together regulators, industry practitioners and regulatory technology vendors from around the world.

After meeting existing clients, speaking to some of the world’s largest financial institutions, and sharing our expertise across a series of roundtable discussions, we’ve pulled together our thoughts on some of the hot topics of discussion.

The volume of eComms false positives are a major problem for firms

It won’t be much of a surprise that eComms surveillance was one of the most talked about regulatory challenges of the show. The huge surge in enforcement action seen in the US this year, coupled with the FCA’s very public announcement that this a key focus area for them has clearly garnered the attention of regulatory professionals on both sides of the Atlantic.

So, while awareness of the challenge is certainly not a problem, it appears that finding an operational solution is. After speaking with a number of large firms during the event, there appears to be a common thread - how do you reduce the massive volume of false positives generated by multiple eComms channels? Several firms independently told us that the sheer number of alerts were swamping their compliance teams, putting a strain on resources, and ultimately making it more difficult to review and identify messages that genuinely require further investigation.

How eflow helps: Our TZTS Trade Surveillance and TZEC eComms Surveillance systems feature highly configurable alert parameters that can be configured in line with your firm’s specific risk strategy. As part of your system, you will also have access to an independent sandbox environment where alert parameters can be tested, refined and then promoted to your live system, enabling you to identify the perfect set up for your team.

Can AI help identify where thresholds should be based upon model feedback loops?

This was a topic that eflow explored as part of the FCA’s recent Market Abuse Surveillance Tech Sprint. The use of an alert feedback loop allows machine learning to understand what should be considered a high or low value alert. As a result, we can then understand the commonalities within these results and apply them to the design of alert logic.

For example, this could be the time traded before the release of Material Non-Public Information (MNPI) for an insider dealing alert, or the time for cancellations of a non-bona fide order for spoofing. These patterns will highlight the type of activity that a firm will consider to be high risk, which can then be applied to both risk scoring and the fine tuning of specific thresholds.

How eflow helps: Our work as part of the FCA Tech Sprint is already being developed and enhanced for use across eflow’s products in the near future. We expect to make further announcements on the integration of this functionality during 2025.

Can AI-generated risk score alerts help analysts get to their risk quicker?

Perhaps unsurprisingly, the use of AI within regulatory technology remains a hot topic. However, while the last couple of years has seen an unrelenting ‘buzz’ around AI and the seemingly endless possibilities it offers, it does feel like the conversation has become a little more ‘realistic’ at this year’s event.

A recurring theme was around the use of AI-generated risk scoring of alerts and whether they can help compliance teams to identify risks quicker. This dovetailed nicely with one of the themes from the FCA’s Tech Sprint which eflow was invited to contribute to. Our opinion is that AI can play a vital role in helping analysts to identify the high value alerts that require higher amounts of attention, rather than having to sift through dozens of alerts to get there. We also believe that machine learning will help to understand the parameters that sit behind these alerts and offer suggestions on how they could be dynamically adjusted over time.

While there is no doubt that AI can (and already is) enhancing operational efficiency in a number of ways, it is important to remember that nothing can fully replace the expertise of an experienced regulatory professional. One should also always remember that it’s the firm and individuals who will be on the receiving end of any enforcement action, rather than the technology.

How eflow helps: Our products use the latest technological advancements to aid operational efficiency wherever possible. In terms of AI, this is most evidently deployed in our TZEC eComms Surveillance system, which uses machine learning to identify linguistic trends at both a company and industry-wide level that are indicators of suspicious behaviour.

A holistic approach to surveillance is high on firms’ agendas

Holistic surveillance is not a new topic, but it remains a key challenge that many firms have yet to find a compelling solution to. Specifically, we found that firms were keen to explore how technology could be used to leverage multiple data points to cut down on the volume of false positive alerts and improve investigation synergies.

How eflow helps: Our integrated trade surveillance solution combines a firm’s structured trading data with unstructured communications data to provide ‘one true view’ of your market abuse risks. In doing so, it provides highly contextualised information that allows firms to easily link digital communications to suspicious trading activity and uncover previously inaccessible insights. This data is then ‘learnt’ by the system so that the use of similar language or terminology in the future is flagged for further investigation before any potentially non-compliant activity has taken place.

Risk assessing your programs for digital assets

With digital assets once again at the forefront of people’s minds, ensuring that firms are MiCA ready is vital. With Title VI of MiCA requiring trade surveillance to be in place by 30th December 2024 (and grandfathering not possible for Trade Surveillance) CeFI exchanges with central limit order books and those offering advice on, and trading with, client funds will need to be compliant soon.

How eflow helps: By offering a trade surveillance platform capable of handling large volumes of data 24/7, together with MiCA and ESMA/MAR- ready alert typologies, we are well placed to understand and mitigate your digital asset risk profile. It’s important to understand that with different tokens and liquidity profiles come different trading risks; eflow can create contextual parameters that avoid compliance teams being swamped with false positives and make sure they get to their risk quicker.

Calibrating your trade surveillance strategy for multiple asset risk profiles

It would be naive to think that a ‘one size fits all’ approach to trade surveillance is likely to meet either the demands of regulators, or the operational needs of firms dealing in multiple asset classes. So the question facing many firms is how do you test and calibrate models for different asset risk profiles and subsequent thresholds?

How eflow helps: eflow’s TZTS Trade Surveillance system features conditional parameters as part of its core functionality, providing firms with the ability to set different conditions for alerts for various types of trading activity. The parameters can be set for factors like market volatility, liquidity, or client type. For example, this means it can distinguish between activity driven by natural market fluctuations and those which may be suspicious.

Thanks to this level of customisation, TZTS increases the accuracy of alerts and significantly reduces the number of false positives reported. This enables firms to have more confidence that their compliance teams are focusing on genuine threats, while also working more efficiently.

In conclusion…

There’s little doubt that financial institutions are facing regulatory challenges that are becoming more sophisticated and nuanced each year. Based on our conversations, it’s reassuring to see that many firms are taking a proactive approach to these challenges which is directly in line with demands from regulators.

With this in mind, assessing and identifying the technological tools that will help a firm to strengthen their regulatory governance while streamlining increasingly complex processes is vital. For more information on how eflow can support your firm’s regulatory strategy, request a no-obligation consultation call with our team of experts today.