European Commission Confirms AI-Assisted Cyberattack, Several Digital Services Temporarily Disrupted
The European Union's executive body says a cyber campaign detected in October involved the misuse of a publicly available artificial intelligence tool, prompting officials to take several applications offline as a precaution while investigators examine the incident.
BRUSSELS, Nov 20: The European Commission has confirmed that it was targeted in a cyberattack involving the misuse of an artificial intelligence tool, raising fresh concerns about the growing role of AI in digital espionage, cybercrime and attacks against government institutions.
The incident, detected on October 22, forced the European Union’s executive body to temporarily suspend access to several digital applications as a precautionary measure. Most of the affected services have since been restored, although investigations into the attack remain underway.
European Commission spokesperson Thomas Regnier confirmed the incident during a press briefing in Brussels, explaining that the investigation had identified the use of artificial intelligence in the cyber campaign.
Officials said the attack involved the misuse of a publicly available AI tool. However, they did not disclose the identity of the attackers, the exact method used to penetrate the systems or whether any specific organisation was suspected of being responsible.
The Commission said it had not found evidence of a data breach affecting its staff as a whole. It also confirmed that preventive measures were introduced immediately after the incident was identified.
The disclosure has drawn attention to the growing cybersecurity risks associated with artificial intelligence, particularly as sophisticated digital tools become more accessible to individuals and criminal groups.
AI systems can assist legitimate cybersecurity professionals in identifying software vulnerabilities, analysing suspicious activities and improving the speed of incident response. However, the same capabilities can also be misused to support malicious operations.
The European Commission’s confirmation represents a significant development because it is among the first publicly acknowledged cases involving an AI-generated cyber campaign targeting a major European Union institution.
According to officials, the attack was detected on October 22 and confirmed following an initial investigation on October 26. The Commission subsequently began taking protective action and launched a wider examination of the incident.
Several applications were temporarily taken offline to limit potential exposure and prevent further disruption. Most of these services have since returned to operation.
The Commission also maintained contact with the European Union Agency for Cybersecurity, known as ENISA, and its internal cybersecurity service, CERT-EU, as part of the response.
Investigators are examining the circumstances surrounding the attack and assessing whether additional systems may have been affected.
Although officials have not confirmed the precise applications targeted, a report by Politico indicated that the incident disrupted internal administrative services for more than 10 days.
The reported disruptions included systems used for staff-related functions such as salaries, timesheets, leave management, medical reimbursements and translation services.
The Commission has not independently confirmed all the specific operational details reported by the publication.
The incident has highlighted the vulnerability of large public institutions that depend on interconnected digital platforms to manage everyday administrative and operational activities.
Government agencies frequently store and process sensitive information across multiple systems, including employee records, financial documents, internal communications and confidential administrative data.
An attack affecting even a limited number of applications can create significant operational difficulties, particularly when the affected services are connected to essential administrative functions.
The European Commission has not disclosed whether the attackers attempted to steal information, gain long-term access to internal networks or disrupt services for political or financial reasons.
Officials have also declined to identify the individuals or groups believed to be behind the operation.
The absence of confirmed attribution means investigators will need to examine technical evidence, digital traces and the methods used during the campaign before drawing conclusions about its origin.
The use of AI in the incident has nevertheless raised questions about how rapidly cyber threats are evolving.
Artificial intelligence tools can help attackers analyse publicly available information, generate convincing phishing messages, automate certain technical activities and identify weaknesses in software systems.
These capabilities can reduce the amount of specialist knowledge required to carry out some forms of cybercrime.
For example, generative AI can produce realistic emails that imitate the communication style of government officials, corporate executives or trusted service providers. Such messages may be used to persuade employees to reveal login credentials or open malicious attachments.
AI systems can also assist in translating fraudulent communications into different languages, potentially allowing attackers to target organisations operating across several countries.
Another concern involves the automation of vulnerability discovery.
Cybersecurity researchers use specialised tools to examine software for weaknesses that could allow unauthorised access. Similar techniques, when misused, may help attackers identify exploitable security gaps more quickly.
However, the presence of AI in a cyber campaign does not necessarily mean that an entirely autonomous system carried out the operation.
Human operators may still select targets, direct activities and make decisions, while AI tools assist with individual stages of the attack.
The European Commission has not provided enough technical information to establish precisely how much of the operation was automated or which capabilities of the AI tool were used.
The incident comes amid growing international concern about the relationship between artificial intelligence and cybersecurity.
Government agencies and technology companies have warned that AI could increase both the scale and speed of digital attacks while also providing defenders with more effective tools for monitoring and protecting networks.
This creates a complicated security environment in which the same technological developments can benefit both attackers and defenders.
The European Union has been working to strengthen its cybersecurity framework through regulatory measures, technical cooperation and investment in digital resilience.
Its institutions and member states increasingly rely on shared digital infrastructure, making coordination particularly important when responding to cross-border cyber incidents.
CERT-EU plays a central role in supporting European Union institutions, bodies and agencies in identifying and responding to cybersecurity threats.
The agency works with affected organisations to investigate incidents, share technical information and strengthen defensive measures.
The Commission’s response also involved contact with the European Data Protection Supervisor, reflecting the importance of assessing whether personal information may have been exposed.
Under European data protection requirements, organisations must take appropriate steps when security incidents potentially involve personal data.
However, the Commission’s current assessment has not identified evidence of a data breach affecting its staff as a whole.
Officials have not ruled out the possibility that further findings could emerge as the investigation progresses.
The incident also raises questions about the security of publicly available AI systems.
Many advanced AI tools are accessible through online platforms, application programming interfaces and downloadable software. While developers introduce safeguards intended to prevent harmful use, no system can guarantee that its capabilities will never be exploited.
Cybersecurity experts have therefore emphasised the need for stronger monitoring, responsible access controls and improved cooperation between AI developers and security agencies.
Organisations are also being encouraged to strengthen their internal systems rather than relying exclusively on software-based safeguards.
Measures such as multi-factor authentication, regular security updates, restricted access privileges, network monitoring and employee awareness programmes can reduce the risk of successful cyberattacks.
Government institutions handling sensitive information may also need to conduct more frequent security assessments and develop procedures for responding to incidents involving AI-assisted attacks.
The European Commission’s experience demonstrates how disruptions to administrative applications can affect the functioning of even well-resourced public institutions.
The attack also highlights the importance of maintaining backup systems and alternative procedures so that essential services can continue during technical emergencies.
For businesses, the incident provides another warning about the need to evaluate the security of AI tools used by employees and third-party service providers.
Companies increasingly incorporate generative AI into customer support, software development, research and document management. These systems can sometimes access sensitive information or interact with other enterprise applications.
If access controls are weak, an AI application could potentially become an additional entry point for attackers.
Organisations therefore need clear policies governing which information AI systems can access, how their activities are monitored and what security requirements apply to external providers.
The European Commission has not announced a final assessment of the financial or operational impact of the cyber campaign.
Its investigation is continuing, and officials have not disclosed whether the incident resulted in permanent data loss or whether any further applications remain affected.
The case is likely to receive attention from cybersecurity agencies and policymakers as governments assess the implications of AI-assisted digital threats.
It also reinforces the growing importance of developing security systems capable of responding to increasingly automated and sophisticated attacks.
As artificial intelligence becomes more widely integrated into everyday technology, protecting digital infrastructure will require cooperation between governments, software developers, cybersecurity companies and users.
The Commission’s response to the incident will therefore be closely watched as European authorities work to establish the full circumstances of the attack and strengthen safeguards against similar operations in the future.