Cube Scales Globally with Two Strategic Acquisitions

With two significant acquisitions announced within days of each other beginning with – Cube acquires Reg-Room to further extend its regulatory intelligence and horizon scanning capabilities followed 8 days later by Cube acquires global regulatory intelligence businesses from Thomson Reuters  – regulatory intelligence specialist Cube has been busy adding to its portfolio of capabilities.

RegTech Insight

Could you share some background on these two purchases and how the pieces will fit together?

Richmond

We agreed a strategic investment from HG in March 2024 – see Cube and Hg Unite – and that was very purposeful for us in the market. We think there’s a significant need in the RegTech sector for companies to have scale, which is needed to deliver the level of capability – and infrastructure – that will ensure data quality is paramount. The comprehensiveness and accuracy of what we do is critical for us and for our customers. That means you need to have lots of different resources that co-exist alongside the technology, AI, and infrastructure to deliver high-quality solutions that ‘don’t miss.’ We deliver solutions that our customers can trust to meet their regulatory requirements.

RegTech Insight

So, it’s really about achieving scale?

Richmond

That’s really what it comes down to. We’ve enjoyed strong growth as a leading tech platform in regulatory intelligence, and we’ve signed many amazing customers globally. We’ve done very well as a firm already, but that’s just the start of Cube’s journey. We see a market opportunity aligned with our strategy to integrate highly adjacent firms that complement the Cube platform and service. We focus on financial services, banking, asset management, investment management, payments infrastructure, and insurance. These areas are all highly synergistic – both in terms of customer focus, as well as adjacent industries.

Reg-Room and the global regulatory intelligence businesses at Thomson Reuters have fantastic pools of subject matter experts (SMEs), regulatory experts, legal experts, researchers, editors, and journalists who have all been creating huge amounts of content over the last 20 years. That expertise is invaluable for augmenting and improving our solutions, which means we can provide more human-curated and human-in-the-loop services. We can now leverage a huge amount of data and human expertise in regulatory analysis, impact assessments, and summarisations.

This purposeful targeting of the acquisitions we’ve done in quick succession gives us absolute scale and synergy with our product proposition. The new combination brings a wealth of data learning, rich expertise, and accelerated R&D and innovation. The data is crucial, providing highly curated, structured, high-quality information for learning and model training.

This is a powerful combination and will allow to innovate and serve a broader set of global customers. We’ll be able to go faster and deliver truly transformative work. With a thousand customers globally, we now have a critical mass to determine what to solve for tomorrow. That’s not to take away what the team has done until today in terms of delivering a strong platform, but we aim to represent the industry more effectively in the future given our larger customer base. We believe we sit in an unparalleled position in the market.

This might look like everything’s happened very fast at Cube and, yes, we’ve been incredibly busy in the last few weeks, but this is very much part of a long-term strategy. We didn’t just come up with a plan overnight. In terms of the long-term goal, we will have all our customers on the Cube platform.

These acquisitions bring complementary technology and product capabilities. We’re integrating the best features into the Cube platform, focusing on unification. It’s all about one Cube, not multiple products in the market.

RegTech Insight

AI has been part of the Cube story from the beginning. Can you discuss where you see frontier AI technologies – generative AI and LLMs – in the Cube roadmap?

Richmond

We use AI extensively across everything we do, from computer vision and data structuring to NLP for text classification and extraction. We also use machine learning for data capture and in-product applications, improving based on customer interactions. In the world of GenAI, we’ve developed our proprietary language model and use LLMs for summarising regulatory obligations, understanding imperative statements, and identifying key topics in regulations.

We’ve built a proprietary language model for regulatory analysis, which we use extensively. We’re pioneering the delivery of AI that customers can use in many ways. We know that more advanced data learning and generative models will further augment and automate compliance processes, which achieves better compliance and also transforms internal efficiencies.

RegTech Insight

Having effectively digitised the path from Regulation to a set of obligations that can be automatically checked against Internal Policy and Procedure, the next step would seem to be connecting this to production systems and measuring compliance as processes are being executed. How are customers leveraging Cube interoperability with internal systems and controls?

Richmond

This involves automating the mapping of customer obligations to policies, procedures, and controls. We aim to provide granular, risk-based insights and full traceability for customers. The next stage is measuring adherence to control requirements related to obligations and creating industry standards and benchmarks. We can present this data in a way that connects to internal processes, helping customers implement and measure compliance effectively.

We have an API that our customers use. We have solutions today where customers will call our API to be able to work out what the regs are in a customer onboarding process or for the mortgage account opening process. So, we see regs at the infrastructure level as rules that can be consumed by machines. “Decision by Machine” is a key part of our roadmap, and that’s part of how Cube is already being used today.

RegTech Insight

FINRA and FCA both make their regulations machine-readable to facilitate regulatory intelligence, but regulatory harmonisation still feels a long way off. How are you finding working with regulators in other geographies?

Richmond

There are two observations to make here. First, the regulatory environment is very pro-quality solutions and technology that can help improve compliance, so we get a lot of positive engagement, good support, and collaboration from the regulators.

But the second challenge always remains: there isn’t the same join-up and the same way in which regulations are produced, classified, and taxonomised. And alongside this, there are differences of interpretation across jurisdictions.

t’s the role of Cube to enable the customer to be fully informed of what they need to know about, we help to remove that cross-jurisdictional cross-regulator complexity.

This means customers can make the right decision but do it in a fully informed way. Then, work with regulators as effectively as possible, possibly via APIs. But really, it’s about good dialogue and good collaboration.

RegTech Insight

How does Cube handle translation with regulations published in multiple languages?

Richmond

Language support is crucial, given the variety of languages regulations are published in. We use multiple translation engines and our proprietary models to ensure high-quality translations. About 7,000 languages are spoken worldwide, but fortunately, only about 70 languages are used for publishing regulations. We have built models around each of these 70 languages to ensure accurate translations. Without consistency and accuracy in translation, everything else falls apart.

We use different engines that are best at handling the core languages, and we overlay our proprietary capabilities to ensure high-quality translation. This process includes our team of 25 regulatory linguists who help train these models and ensure they can detect and interpret regulatory language nuances. Accurate translation is fundamental because it drives the classification of regulations and their applicability, making consistency and accuracy critical for our solutions.