Sometime in September, an attacker used a compromised login belonging to a small Danish IT company to access Denmark’s Central Person Register — the CPR. The system holds names, addresses, and CPR numbers for every person registered in Denmark: living residents, people who have emigrated, and the deceased. The access ran for roughly ten days before anyone noticed.
By the time the breach was disclosed on October 5, the records of approximately 8.8 million people had been exposed.
Denmark’s population is about 5.9 million. The register includes historical records for emigrants and the dead, which explains why the number of exposed records exceeds the current population. Nearly everyone who has ever been legally registered in Denmark was affected.
One of the account passwords used was “123456.”
What Is the CPR System?
The Central Person Register is not just a database — it is the foundation of Danish civic infrastructure. A CPR number is used for everything: healthcare, banking, taxes, social services, school enrollment, and identification. It functions similarly to a Social Security number in the United States but is more deeply embedded in daily life. Without a CPR number, you cannot open a bank account, see a doctor, or register a child for school.
The system is centrally administered and accessed by a wide range of authorized parties — hospitals, employers, government agencies, and private companies that need to verify identity for regulated services. That breadth of authorized access is precisely what made the breach possible.
How It Happened: A Third-Party Access Failure
The attacker did not break into the CPR system directly. They obtained credentials belonging to an authorized third party — a small company called Pays, based in the Funen region — and used that company’s legitimate access to extract data from the register.
Datatilsynet, Denmark’s data protection authority, described the attack as involving some form of credential abuse combined with enumeration: the attacker used valid CPR numbers (which have a predictable structure) to query the register systematically and extract corresponding records at scale.
The Politiken newspaper reported that at least three user accounts at the company used the password “123456.” It is not yet confirmed whether the initial compromise exploited one of those accounts specifically, but it illustrates the authentication environment the attacker was working with.
The access was blocked once discovered, and Danish police opened an investigation. The perpetrator has not been publicly identified.
Why the “Third-Party Access” Vector Keeps Appearing
The CPR breach is the latest in a long series of high-profile incidents that entered through the same door: a vendor, contractor, or service provider with legitimate access to sensitive data, operating with inadequate security controls.
In the DHS Icepick breach disclosed on October 28, initial reporting pointed to Okta — a single sign-on provider that legitimate employees use to authenticate. In the 2023 MOVEit campaign, attackers compromised a widely-used file transfer tool and used that access to reach data at hundreds of organizations that had nothing wrong with their own security posture. In the 2024 Snowflake-related incidents, credentials for a cloud data warehouse used by dozens of large companies were compromised, giving attackers access to those companies’ customer data without ever touching the companies’ own systems.
The pattern is consistent: the organization that holds the sensitive data invests in its own security, but the supply chain of entities with authorized access to that data is far harder to audit and control. An attacker who cannot get through the front door may be able to walk in through a contractor’s entrance.
For the CPR system, the implication was stark: the central registry was presumably protected. The small company with legitimate access to it was not.
What Data Was Exposed and What Was Not
The CPR register stores more information than names, addresses, and ID numbers — it includes relationship records, citizenship data, and historical address histories. Reports from Datatilsynet indicate that the exposed data includes names, current addresses, and CPR numbers, but not records that are subject to name-and-address protection. Denmark allows individuals to shield their address from public lookup for safety reasons, and those protected records reportedly were not accessed.
This distinction matters: for people in the name-and-address protection category — domestic violence survivors, witnesses, people with stalkers — their protected status held. For the approximately 8.8 million records that were not protected, full identification data is now in an attacker’s hands.
The exposed data does not include financial information or health records, but a CPR number combined with a full name and address is enough to:
- Impersonate someone in contexts that accept CPR-based identity verification
- Build convincing phishing attacks tailored to specific individuals
- Combine with other leaked datasets to create more complete profiles
- Potentially open financial accounts in services that accept CPR as a primary identifier
The Centralization Problem
The CPR breach illustrates the fundamental tension in building modern civic infrastructure: centralization is efficient, but it creates catastrophic single points of failure.
A decentralized identity system — where each agency holds only the minimum data it needs, with no single repository that combines all of it — would have limited the blast radius. An attacker who gained access to the healthcare registry would have names and health records but not addresses or CPR numbers from the same query. An attacker who accessed the banking registry would have financial identifiers but not health or civic data. Correlation would require breaching multiple systems and combining the results, which is substantially harder.
Denmark’s CPR system was designed for a different threat model — one where the primary risks were administrative errors and unauthorized internal access, not systematic credential compromise through third-party vendors. The threat model changed faster than the architecture did.
What This Means for Personal Data Storage Generally
Most people reading this are not Danish residents and do not have CPR numbers. The lesson is not Denmark-specific.
The relevant question the breach poses is: what happens to your most sensitive personal information when it lives in a system you did not build and cannot audit?
For national ID registries, the answer is provided by the state, and individuals have limited ability to opt out. But for the personal data you store voluntarily — documents, photos, health records, financial files, personal notes — you have meaningful choices about where it lives, who can access it, and what happens when that system is compromised.
The CPR breach involved information that victims had no choice about storing in a central registry. Most personal file storage involves information people have chosen to put in a specific cloud service. That choice comes with different risk profiles for different services:
- Services that aggregate your data across products (calendar, email, photos, documents) create a profile that is far more valuable to an attacker than any single piece of data
- Services that rely heavily on third-party vendors — analytics, customer support tools, cloud infrastructure — inherit those vendors’ security posture whether they want to or not
- Services that grant authorized third-party access (developers, integration partners) extend the attack surface to every party in that ecosystem
Minimizing the data you hold in centralized systems and keeping your most sensitive files in a dedicated, purpose-built storage layer reduces your exposure to this class of breach without requiring you to predict which specific system will fail next.
Practical Steps After a Breach of This Scale
For people affected by the CPR breach, or for anyone thinking through the implications of large-scale identity data exposure, the most useful responses are:
Monitor for targeted phishing. Attackers with name, address, and ID number can craft highly personalized messages. Any communication that references personal details and requests urgent action — payment, credential verification, security updates — deserves extra scrutiny regardless of how official it looks.
Check credit and identity monitoring services. Most Scandinavian banks offer monitoring for credit applications in your name. Enabling alerts costs nothing and can catch misuse early.
Understand what cannot be changed. A CPR number, like a Social Security number, cannot simply be reissued to address a breach. The protective steps focus on making the exposed data harder to exploit, not on undoing the exposure itself.
Separate high-risk files from routine cloud storage. The documents and records with the highest identity-theft risk — passports, residency documents, financial statements, medical records — are worth storing in a dedicated private vault rather than in the same cloud folder as grocery lists and screenshots.
The Long Arc of Centralized Identity
The Denmark CPR breach will not be the last incident of this kind. The architecture of centralized national registries was designed before the current threat landscape existed, and it is being tested by attacks the original designers did not anticipate.
The response — better credential controls, stricter third-party access audits, multifactor authentication for systems that touch sensitive registries — is necessary but reactive. It addresses the specific failure mode that the September incident exploited without changing the underlying reality: concentrating eight million people’s identification data in one queryable system creates a target that is worth attacking at considerable effort.
For individuals, the practical lesson is to treat large centralized repositories of personal data — national registries, major cloud platforms, employer HR systems — as systems that can be reached, even when you cannot see how. What you choose to store in those systems, and what you keep elsewhere, is one of the few aspects of this risk that individuals can actually control.
daftei keeps personal files encrypted in transit with TLS 1.3 and at rest with AES-256, with no advertising, no data sales, and no third-party AI training on user content. 5GB of storage is free; Pro offers unlimited storage for $5.99/month, $44.99/year, or $89.99 lifetime ($89.99 one-time). In India, Pro is ₹249/month or ₹1,799/year. daftei is GDPR and CCPA compliant. Account deletion initiates a 30-day grace window, after which all data is permanently and irreversibly erased.