Novo Nordisk Confirms Cyberattack Exposing Patient Medical Data and Internal AI Assets

Novo Nordisk, the Danish pharmaceutical giant behind blockbuster weight-loss drugs Ozempic and Wegovy, has confirmed a cybersecurity breach involving unauthorized access to sensitive clinical data and internal artificial intelligence (AI) assets.

The company disclosed that attackers successfully exfiltrated a limited volume of non-public information, raising concerns over data security within highly regulated healthcare and research environments.

Novo Nordisk Cyberattack

According to a post from Novo Nordisk on X, the compromised data includes patient-related attributes such as unique identifiers, sex, birth year, biomarkers, and health indicators, as well as lifestyle-related factors such as body mass index (BMI) and smoking status. Additionally, contact information of certain healthcare professionals was exposed.

However, the company emphasized that no direct identifiers, such as patient names, addresses, or national identification numbers, were accessed, reducing the likelihood of immediate identity theft.

The organization stated that its core infrastructure, including drug production, supply chain operations, and clinical trial processes, remains unaffected.

Novo Nordisk has engaged external cybersecurity experts to investigate the incident and has notified relevant regulatory authorities. The company currently assesses the risk to affected individuals as low but continues to monitor the situation closely.

Threat intelligence sources, including vx-underground, suggest that the threat actors behind the breach are attempting to extort victims by leaking samples of allegedly stolen data.

These samples point to a potentially deeper compromise involving Novo Nordisk’s internal AI ecosystem. Among the claimed exfiltrated assets are a 16.7 GB trained model checkpoint, a 407 MB proprietary training dataset, and full source code tied to internal AI development projects, including files such as “modeling_novopert.py” and associated training pipelines.

Further claims indicate exposure of 113 AI training runs with detailed logs, internal infrastructure mappings covering HPC clusters, Slurm workload managers, and SSH configurations, and over 53 GB of container images.

The threat actors also allege access to developer identities, internal hostnames, and private GitHub repository URLs, suggesting a potential compromise of development environments and software supply chains.

Notably, Novo Nordisk has not officially confirmed the authenticity of these AI-related data leaks. However, if validated, the exposure could represent a significant intellectual property risk, particularly given the competitive value of proprietary AI models and biomedical datasets in pharmaceutical research.

Security researchers have also speculated that AI may have assisted elements of the attack, although this remains unverified. Such a scenario would align with the growing trend of adversaries leveraging AI tools to automate reconnaissance, enhance phishing campaigns, or optimize lateral movement within compromised networks.

This incident highlights the increasing convergence of healthcare data security and AI infrastructure protection, where breaches can impact not only patient privacy but also critical research assets.

As investigations continue, the attack underscores the need for robust security controls across both clinical data repositories and AI development pipelines in the pharmaceutical sector.

Follow us on Google News, LinkedIn, and X to Get Instant Updates and Set GBH as a Preferred Source in Google.