Explore our new Cybersecurity Terms Glossary, explained by our experts.

View the Glossary View the Glossary
PRIVIA

Cybersecurity Terms Glossary

What Is the SWIFT Customer Security Programme?

The SWIFT Customer Security Programme requires every institution on the payment messaging network to implement a defined set of controls and attest to them annually, with independent assessment.

Illustration for the cybersecurity term SWIFT Customer Security Programme
Home / Blog / Cybersecurity Terms / What Is the SWIFT Customer Security Programme?
17 November 2026

What Is the SWIFT Customer Security Programme?

What Is the SWIFT Customer Security Programme?

The SWIFT Customer Security Programme, usually shortened to CSP, is the security regime that institutions must follow to stay on the network that carries interbank payment instructions. It defines a set of controls covering the local environment where payment messages are created and sent, requires each institution to attest annually to its compliance, and requires that attestation to be backed by an independent assessment. Attestations are visible to counterparties, which turns compliance into a commercial matter as well as a technical one.

The programme exists because of a theft. In February 2016, attackers inside Bangladesh Bank issued fraudulent payment instructions through the bank’s own SWIFT terminal, attempting to move around 951 million US dollars from its account at the Federal Reserve Bank of New York. Roughly 81 million dollars reached casinos in the Philippines and most of it was never recovered. The network itself was not compromised; the customer’s local environment was.

What the Framework Covers

The Customer Security Controls Framework is organised around three objectives and a set of principles beneath them. In outline:

The framework is revised annually, with controls periodically promoted from advisory to mandatory, so an institution that met the requirements two years ago may not meet the current version. Each institution maps itself to an architecture type, from a full local installation through to a relationship entirely mediated by a service bureau, and the applicable controls follow from that type.

Attestation and Assessment

Every user attests annually against the current framework version, recording compliance or non-compliance per control with any exceptions explained. Since the 2021 cycle, the attestation must be supported by an independent assessment, carried out either by an internal audit function that is independent of the operational team or by an external assessor. Self-attestation without assessment is no longer acceptable.

Counterparties can read attestation data, and increasingly do. A correspondent bank that sees unresolved non-compliance may restrict or reprice the relationship, which is why the programme produces action in institutions that regulatory pressure alone had not moved.

Making It Work in Practice

  1. Draw the boundary honestly. The secure zone contains the messaging interface, the communication interface, connectors, jump servers, and the operator workstations, and the common failure is a workstation that is also used for email and browsing.
  2. Harden the operator path. Dedicated or virtualised workstations with no general internet access, multi-factor authentication, and no local administrative rights.
  3. Enforce four-eyes. Message creation and approval must be separate people with separate credentials, with the control applied in the application rather than in procedure.
  4. Use the network’s own detection tools. Daily validation reports and payment controls that block or hold messages outside expected patterns catch the fraudulent instruction that slipped through everything else.
  5. Protect and externalise logs, because attackers in historical cases tampered with local records and printers to delay discovery.
  6. Reconcile independently. The Bangladesh Bank attackers manipulated confirmation output to hide the transfers; reconciliation against a source the attacker does not control is what shortens the detection window.
  7. Include service bureaux and group hubs in the assessment, since the obligation follows the relationship, not the hardware.

Where It Sits Against UK Regulation

The CSP is a contractual requirement from a member-owned cooperative, not a law. For a UK bank it sits alongside the PRA and FCA operational resilience rules, outsourcing expectations in Supervisory Statement 2/21, and, since the start of 2025, the critical third parties regime created under the Financial Services and Markets Act 2023. The Bank of England also runs CBEST threat-led penetration testing for major firms, and payment messaging infrastructure is a natural scenario within it.

The overlap is substantial but not complete. CSP controls apply narrowly to the payment messaging environment and are prescriptive; UK operational resilience applies broadly and is outcome-based. Institutions that treat the annual attestation as a compliance exercise separate from their wider security programme generally end up maintaining two sets of evidence for the same controls. Those that map CSP controls into their existing framework produce one.

Our regulation and compliance services support attestation readiness, including independent assessment of the secure zone and technical validation of segregation and access controls.

FAQ – Frequently Asked Questions

Is CSP Compliance Legally Required?

It is contractual rather than statutory. The requirement comes from the cooperative that operates the network, and the consequence of failure is loss of access or reporting to the relevant supervisor, which in practice is as compelling as a legal obligation. UK regulators expect firms to manage the risks in their payment infrastructure regardless, so the two reinforce each other.

Can We Attest Without an Independent Assessment?

No. Since the 2021 attestation cycle, every attestation must be supported by an independent assessment, performed either by an internal audit or compliance function that is independent of the team operating the environment, or by an external assessor. The assessor must have relevant expertise, and the assessment must cover the controls applicable to the institution’s architecture type.

Does CSP Apply If We Use a Service Bureau?

Yes, with a different control set. Architecture type determines which controls apply, and institutions whose connectivity is entirely mediated by a service bureau or a group hub still attest, focusing on operator access, credential protection, and the controls they retain. The bureau’s own compliance does not substitute for yours, and its attestation is worth reviewing as part of supplier assurance.

What Was the Core Failure at Bangladesh Bank?

The local environment, not the network. Attackers obtained operator credentials, operated inside a poorly segmented environment, used custom malware that interfered with the messaging application’s checks, and manipulated confirmation output and printing to delay detection over a weekend. Every major element of the control framework maps to one of those failures.

Related Terms

You May Be Interested In These

What Is Hypothesis-Driven Threat Hunting?

Cybersecurity Terms

17 November 2026

What Is Hypothesis-Driven Threat Hunting?

Read More
What Is the Traffic Light Protocol?

Cybersecurity Terms

17 November 2026

What Is the Traffic Light Protocol?

Read More
What Are Azure Managed Identities?

Cybersecurity Terms

17 November 2026

What Are Azure Managed Identities?

Read More