Skip to main content
Promotional banner for the pentest readiness checklist
Sweden's $183,000 Fine Shows What Happens When You Skip Software VettingBreach & Risk Assessment
4 min readFor Legal and Compliance Teams

Sweden's $183,000 Fine Shows What Happens When You Skip Software Vetting

What Happened

On August 25, 2025, Miljödata, a Swedish IT systems provider serving 80% of Sweden's municipal work environment and HR systems, suffered a cyberattack that compromised sensitive data for 2.2 million people. The attacker demanded 1.5 Bitcoin (worth $168,000 at the time) and, when payment wasn't made, published the stolen data on the dark web under the name "Datacarry."

The exposed information included personal identity numbers, contact details, sickness absence records, rehabilitation data, and school incidents involving minors. Sweden's supervisory authority, IMY, launched an investigation in November 2025 and concluded that Miljödata violated GDPR Article 32(1) by failing to maintain adequate security measures. The agency imposed a fine of $183,000 (SEK 1.8 million).

Timeline

  • August 25, 2025: Cyberattack disrupts IT services across 200+ regions
  • Shortly after breach: Attacker demands 1.5 Bitcoin ransom
  • After demand ignored: Data published on dark web
  • November 2025: IMY opens formal investigation
  • 2025 (date not specified): IMY issues $183,000 penalty; investigations into two municipalities and one region remain ongoing

Which Controls Failed or Were Missing

IMY identified two specific failures in Miljödata's security posture:

1. Inadequate software vetting procedures

Miljödata didn't perform sufficient checks when installing new software. This matters because threat actors often compromise supply chains or exploit vulnerable software components. Without a structured vetting process, including code review, vulnerability scanning, and vendor security assessments, you're essentially trusting that every piece of software you deploy is safe.

2. Absence of automated real-time monitoring

The company lacked automated mechanisms to detect intrusions and suspicious activity as they occurred. This isn't just about having logs; it's about having systems that analyze those logs continuously and alert you when something deviates from normal behavior. Real-time monitoring catches lateral movement, privilege escalation, and data exfiltration while you still have time to contain the damage.

These weren't theoretical gaps. The breach affected 2.2 million people and exposed some of the most sensitive categories of personal data: health information, data about minors, and government identification numbers.

What GDPR Article 32(1) Requires

Article 32(1) requires controllers and processors to implement "appropriate technical and organisational measures to ensure a level of security appropriate to the risk." The regulation specifically mentions:

  • Pseudonymization and encryption of personal data
  • The ability to ensure ongoing confidentiality, integrity, availability, and resilience of processing systems
  • The ability to restore availability and access to data after an incident
  • A process for regularly testing and evaluating the effectiveness of security measures

The phrase "appropriate to the risk" is critical. Miljödata processed health data and information about children, both classified as special categories under Article 9. The company also served 80% of Sweden's municipal systems, creating systemic risk. That context demands controls beyond basic security hygiene.

IMY's conclusion was direct: Miljödata "did not maintain a sufficiently high level of technical and organizational security, considering the types of personal data it processed."

Lessons and Action Items for Your Team

Stop treating software deployment as a technical formality

Create a software acceptance procedure that requires:

  • Documented security review before production deployment
  • Vulnerability scanning against current CVE databases
  • Vendor security questionnaires for third-party components
  • Approval from both IT and information security teams

If you're deploying software that will process special category data, add penetration testing to that list.

Deploy detection, not just prevention

Prevention controls fail. Your monitoring system should:

  • Analyze authentication patterns and flag anomalies (failed logins from unusual locations, access outside normal hours)
  • Track lateral movement between systems
  • Alert on bulk data access or export operations
  • Monitor privileged account activity in real time

These alerts need to reach someone who can act on them. If your security team learns about suspicious activity from a user complaint rather than from your monitoring system, you don't have real-time detection.

Document your risk assessment for high-risk processing

Article 32(1) requires security "appropriate to the risk." If a supervisory authority challenges your measures, you need to show that you assessed the risk and made informed decisions. Document:

  • What categories of data you process
  • The volume of data subjects affected if a breach occurs
  • The potential harm to individuals (reputational, financial, physical)
  • Why you selected specific controls over alternatives

This documentation also helps during Prior Consultation under Article 36, if your processing meets the threshold for mandatory consultation.

Understand the economics of ransom demands

The attacker in this case demanded $168,000, less than the eventual regulatory penalty. Sophisticated threat actors research potential regulatory exposure and set demands below that threshold to make payment seem rational. Don't assume a low ransom means low impact. The regulatory penalty is only one component of breach costs; add notification expenses, system restoration, legal fees, and reputational damage.

Prepare for downstream investigations

IMY noted ongoing investigations into two municipalities and one region connected to the Miljödata breach. If you're a processor, your breach creates compliance exposure for every controller you serve. Your contracts should specify:

  • Notification timelines to controllers after you detect an incident
  • Your obligations to support controller breach notifications
  • Liability allocation for regulatory penalties

If you're a controller relying on processors, Article 28 requires you to verify that they provide "sufficient guarantees" of security. You can't outsource accountability.

The Miljödata case demonstrates that supervisory authorities will examine your security architecture in detail after a breach. Generic statements about "industry standards" won't satisfy them. They want to see that you identified your specific risks and deployed controls matched to those risks. If you process sensitive data at scale and you can't explain why you don't have real-time monitoring, expect that conversation to be expensive.

Promotional banner highlighting failures found in PCI audits and how to spot the gaps

You Might Also Like