CASE STUDY: TACKING THE MULTI-CLOUD COMPLIANCE TAX

Progress from RBC, Morgan Stanley, and FINOS Common Cloud Controls

The Challenge: Multi-Cloud Compliant Tax

Cloud services represent critical infrastructure for modern financial institutions, yet operating across multiple cloud service providers (CSPs) introduces significant compliance and operational friction. When global financial institutions deploy workloads across Microsoft Azure, Amazon Web Services (AWS), and Google Cloud Platform (GCP), security and risk teams are forced to answer thousands of repetitive compliance questions.

Across the industry, multi-cloud assurance historically suffers from two structural fractures:

  • The Tool-to-Control Definition Gap: Because cloud providers implement configurations differently, engineering and security teams often focus more on writing vendor-specific rules (such as Azure Policy or AWS Config) than on the actual security control. When controls are implemented across disparate cloud mechanisms, security gaps inevitably emerge.
  • The Spreadsheet & Evidence Gap: Compliance baselines are traditionally documented in static, internal spreadsheets. Engineers are handed these documents and tasked with independently scripting infrastructure patterns (e.g., Terraform) for each cloud provider. Consequently, organizations end up with multiple disconnected implementations, no standardized way to validate deployed infrastructure against original policies, and immense difficulty proving compliance to regulators against external benchmarks such as NIST.

The Solution: A Shared, Machine-Readable Framework

FINOS Common Cloud Controls (CCC) provides an open, vendor-agnostic framework that establishes a shared language for cloud security and governance. Rather than introducing another proprietary checklist, CCC organizes cloud security across five core components:

  • Capabilities: Defining the common functionalities of cloud services across AWS, Azure, and GCP in an agnostic manner.
  • Threats: Identifying specific risks and attack vectors associated with those capabilities, mapped to frameworks such as MITRE ATT&CK.
  • Controls: Establishing preventive and detective measures to mitigate identified threats, mapped directly to external standards like NIST.
  • Assessment Requirements: Defining cloud-neutral criteria used to verify that a control is actively met.
  • Validation: Executing automated behavioral test suites within CI/CD pipelines to verify live cluster configurations.

In 2026, the CCC Steering Committee was refreshed with a renewed purpose, uniting cloud leads and cybersecurity executives from key financial and technology leaders—including Citi, LSEG, Morgan Stanley, RBC, BlackRock, Red Hat, and Scott Logic.

2026-11 Case Study - CCC Steering Committee image

In this case study, we focus on the progress made in 2026 by two of these firms: RBC and Morgan Stanley. Additionally, broad progress has been made across the wider project, which we recap at the end of the case study.

Upstream Governance: RBC’s Separation of Intent from Enforcement

"Sometimes we focus more on the policy creation versus on the control... when you have different implementations, you can have gaps and you need to have a framework on top of that to be sure all these gaps are covered. And that’s why CCC will help us have a high-level, agnostic control catalogue.”

— Maxime Coquerel, Principal Cloud Security Architect, RBC

Maxime Coquerel, Principal Cloud Security Architect at RBC and CCC Security Working Group Lead, took to the stage at OSFF Toronto in March 2026 to talk about his firm’s use of Common Cloud Controls. There, he pointed out the risk of focusing too heavily on provider-specific policy syntax rather than the underlying security objective.

He shared how requirements can be defined once at the capability layer rather than reinvented per tool. For example, the requirement mandating TLS encryption in transit for Object Storage is defined once in CCC. Technical enforcement is then delegated natively to Azure Policy on Azure Storage Accounts and to AWS Config on Amazon S3, maintaining consistent security intent across clouds.

This session is available on YouTube at: https://youtu.be/rZZGisQBzcQ

“By mapping CCC’s cloud-agnostic security control objectives to corresponding controls in RBC’s internal Cloud Security Framework, the bank establishes a clear connection between industry-standard security requirements, internal policies, and cloud-specific technical implementations. This mapping enables security teams to identify which internal controls address each CCC objective, reuse existing security assessments, and maintain traceability across cloud providers. As a result, RBC can accelerate security reviews, reduce duplicated effort, and improve consistency while ensuring alignment with internal security standards and regulatory expectations. Ultimately, CCC provides a shared security language that connects security architects, engineering teams, and governance stakeholders, supporting more efficient and scalable cloud security governance.”

— Maxime Coquerel, Principal Cloud Security Architect, RBC

Downstream Platform Assurance: Morgan Stanley’s Perspective on Managed Kubernetes

"If Morgan Stanley has this problem, and we do have this problem, right? I guarantee every other financial services organization is going to have exactly the same problem."

— Dave Reeve, Morgan Stanley

 Addressing the infrastructure side of the equation at OSFF London in June 2026, Dave Reeve (Cloud Lead, Morgan Stanley) highlighted the downstream platform dilemma facing organizations scaling complex compute services across clouds.

Managing enterprise container infrastructure requires organizations to define internal control requirements, such as private API endpoints and modern cryptographic ciphers, mapped to internal policies and NIST standards. However, Reeve detailed how executing these requirements across clouds creates operational fragmentation:

  • Separate engineering teams are often tasked with independently building clusters across Azure AKS, AWS EKS, and Google Cloud GKE.
  • This produces three disparate implementations with no standardized way to validate whether they meet the same controls or maintain an identical security posture across providers.
  • The underlying policy remains locked in static spreadsheets, leaving no automated chain of evidence connecting deployed infrastructure back to regulatory standards.

To resolve this, Reeve outlined an approach anchoring managed Kubernetes strategies to CCC:

  • Machine-Readable Baseline: Translating internal policy definitions into CCC's shared language to establish a direct chain of custody from internal policy down to infrastructure blueprints.
  • Open Reference Architecture: Utilizing open-source Terraform reference implementations across AKS, EKS, and GKE that meet stringent CCC control baselines, allowing firms to deploy hardened patterns out of the box.
  • Continuous Behavioral Validation: Replacing subjective audit assertions with automated behavioral test suites (e.g., Cucumber-based tests) that run against live environments in AWS, Azure, and GCP. These tests actively verify that guardrails function, providing objective proof that "all the doors are locked".

This session is available on YouTube at https://youtu.be/NHvf31wJ2Ig

Breaking Changes at OSFF New York - November 2026

Since these community presentations, work on Common Cloud Controls has accelerated, turning these architectural goals into concrete, shipped deliverables. Organizations do not need to overhaul their entire operating model at once to benefit from this progress; member firms can adopt the framework modularly, whether upstream in architecture and risk governance, downstream in deployment validation, or across both.

Key milestones and opportunities now available to the community include:

  • The CCC.K8S Catalog: The CCC managed Kubernetes catalog has published capabilities, threats, and controls for AKS, EKS, and GKE, providing the shared, machine-readable baseline discussed at OSFF.
  • Updated versions of existing 12 Catalogs: From Object Storage, Secrets, RDBMs, Identity and more, each consisting of sub-catalogs containing capabilities, threats and controls.
  • Hardened Open-Source Blueprints: Moving from theory to code, Yash Gandhi (Morgan Stanley) recently contributed a hardened, CCC-oriented AKS Terraform blueprint to the Azure Verified Modules (AVM) project.
  • Automated Continuous Verification: Multi-cloud behavioral tests in the CCC repository are now actively running against AWS, Azure, and GCP, checking live clusters against the controls to create an automated chain of evidence.
  • Unified Regulatory Assurance: Firms can now present compliance officers and regulators with consistent control baselines tied directly to industry benchmarks like NIST and CRI.
  • Driving Vendor Feature Parity: The financial industry is increasingly equipped to speak with "one voice" to CSPs, using reproducible open-source test cases when cloud security capabilities differ or fail.
  • Managed Kubernetes Behavioural Tests: Tests written to exercise CCC’s assessment requirements against running cloud infrastructure, closing the evidence gap loop between the specifications and live environments.

CCC Morgan Stanley RBC Case Study Vertical Cover

Learn More & Get Involved

Visit the FINOS CCC project site to watch the introductory video, explore resources including catalogs and tools, participate in working roups, and contribute code and catalogs.

FINOS CCC: ccc.finos.org

FINOS Calendar: finos.org/calendar