Date:
The Compliance Burden in Financial Technology
Standards and laws governing regulatory compliance in the financial space continuously evolve across jurisdictions. At the same time, rules and requirements may overlap or simply be highly extensive in coverage and numbers of articles. Financial institutions operate within a complex regulatory environment where legal obligations, risk controls, technology requirements and compliance processes are deeply interconnected.
At its heart, regulatory compliance is a translation problem. Regulations rarely specify exact thresholds or mechanisms but rather provide vague requirements for risk controls and/or operational resilience. It is left to an institution to translate that intentionally general language into working software, and to prove the translation holds.
For decades, that job belonged to lawyers and compliance officers. Policies became PDFs, controls became spreadsheets, and audits became expensive reconstruction exercises – technology came last, so the systems doing the real work could be only loosely connected to the obligations behind them. Every time a rule book changes, that gap widens further: links between obligations and test evidence go stale, until teams can no longer say with confidence that every obligation has been translated into a verifiable requirement, implemented correctly, and backed by sufficient evidence.
This complexity makes AI a valuable automation use case for the compliance function – it can streamline the effort of updating and maintaining regulatory texts, turning legal corpora from static documents into dynamic, data-driven systems with executable controls. But that raises a new question: once AI drafts a compliance check, how do you know it's right? That takes a separate, rule-based layer – one that defines in advance exactly how a result will be verified, independent of how it was produced.
Almost every regulatory requirement comes down to a few basic software concepts: an obligation is a requirement, a threshold is a configuration parameter, evidence is a log. Regulation is simply another specification that software must satisfy – and specifications can be tested. That's what th2-RC does.
Meet th2-RC
th2-RC is a conformance automation framework designed to close this growing gap. A natural next step in Exactpro's evolution, the solution is built on our experience supporting the technology transformation programs of roughly half of the world's top twenty exchanges and exchange groups across twenty-five countries – it extends Exactpro’s proven approach to independent system analysis into a domain where explainable traceability, objective validation and defensible evidence are becoming increasingly critical.
In building it, we made a deliberate choice about where AI belongs in this workflow: it can be used to translate rule book provisions into verification code, but it never touches test execution, and every test result is analysed by deterministic code.
th2-RC Capabilities
-
Aligns rule books, system requirements, testing artefacts and coverage reporting
Bi-directional mapping of obligations and rulebook provisions to linked test evidence
-
Supports any regulatory framework, asset class or jurisdiction
A scalable, reliable foundation for complex compliance obligations
-
AI-enabled where it's safe, deterministic where it matters
AI-assisted coding underpins the process of translating the rules into standardised data structures and event definitions. Test execution and results analysis stay deterministic and auditable.
-
Compliance by Design
Compliance requirements are translated into system requirements, business rules and architecture decisions. Controls, auditability and evidence generation are embedded directly into the software.
The video was created as a Sibos Perfect Pitch 2026 entry.
How th2-RC Fits the Testing Process
The compliance testing flow is complex, but its steps call for different kinds of effort. Sending and receiving transactions is largely mechanical and easy to automate – interpreting transactional data against regulations is a much harder task. That's why th2-RC isolates the flow into separate stages – test generation, test execution, data interpretation – and gives each one the automation approach it needs.
Traditional testing scripts compare one specific expectation to one actual result. That works for functional checks, but regulatory compliance needs more context: the message, the state transitions of the order behind it, the state of the system before and after, and the wider universe of data streams and risk controls. Because compliance lives in the entire system's states and data flows, th2-RC gives that data multiple representations – a Rulebook view tying each provision to its verification logic and coverage, a State Transitions view showing every order's journey through the system and a Messages view that keeps the original traffic one click away.
Behind these views sits the decision that makes the framework versatile: a transactional message isn't just tag-value combinations – each message becomes a categorised signal, and abstract concepts like aggressive pricing or a price-time priority violation get formal definitions in code, adaptable to any financial domain's regulatory landscape.
Recognition along the way
We launched th2-RC in March 2026, at Malta's Governance, Risk, Compliance and ESG Conference hosted by the Malta Stock Exchange and Malta Financial Services Authority.
Earlier in the year, Exactpro won Best Model/Application Testing Environment at A-Team Group's TradingTech Insight Awards Europe, and our innovation efforts resulted in multiple Swift hackathon achievements – Winner in 2022, Runner-up in 2021 and Honorable Mention in 2025. The track record continues this year, with Exactpro’s proposed solution, 'Mobilising Tokenised Collateral Through Standards-Based Lifecycle Management’, selected as a finalist in the Business Challenge track of Swift Hackathon 2026.
Explore how th2-RC can streamline your regulatory compliance reporting – get in touch with our team via info@exactpro.com.