Threat Labs Report: Europe 2026

AI adoption across organizations in Europe has reached near full saturation over the past year, with 99% of organizations now using AI applications. This widespread adoption reflects increasing maturity and confidence in AI technologies, as organizations across Europe continue integrating AI more deeply into their operations and align closely with broader global adoption patterns.

At the same time, user adoption has grown significantly, with the share of users actively using AI applications increasing from 35% to 65%. This trend highlights not only the broad availability of AI tools, but also their deeper and more consistent use across the workforce.

Organizations in Europe have taken significant steps to reduce shadow AI risks by shifting users away from personal AI accounts and toward organization-managed tools. Over the past year, the percentage of people using personal AI applications has dropped from 79% to 43%, while the percentage using organization-managed AI solutions has increased even more sharply from 28% to 72%. However, the number of users switching between personal and enterprise accounts has grown from 7% to 15%, suggesting that organizations still need to match the convenience, accessibility, and features users expect, and deploy more instance controls.

Overall, this shift reflects stronger governance, improved oversight, and a clear move toward managed environments that enhance data protection, compliance, and risk control while continuing to support innovation.

In Europe, the top AI applications show a slightly different pattern compared to broader global trends. ChatGPT remains the most widely adopted AI app, used by 88% of organizations. Notably, Anthropic Claude has strengthened its position in second place at 79%, well ahead of Google Gemini at 69%, marking a clear departure from the more typical global ranking, which currently places ChatGPT in the lead, followed by Google Gemini and then Anthropic Claude. 

This shift highlights growing interest in alternative AI providers and a more diversified adoption landscape across Europe, for example, Mistral Le Chat is now used by 8% of organizations. The remaining leading applications include a mix of specialized and workflow-integrated AI tools that support a wide range of operational and customer-facing use cases.

The chart below illustrates how usage of the top AI applications in Europe has evolved over the past year, highlighting notable shifts in platform preference. During this period, ChatGPT has remained consistently strong.

The most significant shift began in September 2025, when Anthropic Claude saw a rapid increase in adoption, quickly climbing into the second spot among the most widely used AI applications. In contrast, Google Gemini and Microsoft Copilot have remained relatively stable over the same period, showing more gradual and consistent usage patterns.

Overall, these trends point to a diversifying AI ecosystem in Europe, with organizations continuing to rely heavily on established leaders while increasingly embracing alternative AI platforms.

As AI adoption continues to expand across organizations in Europe, concerns around data exposure are being voiced with growing volume. Organizations are using AI for tasks such as summarizing documents, generating reports, and supporting operational workflows: activities that often involve sensitive business and customer data and expand the potential attack surface. As AI becomes more embedded in business processes, data protection remains a top priority, particularly in the presence of ongoing shadow AI risks.

Analysis of data policy violations in Europe shows that regulated data remains the largest category, accounting for 59% of incidents. Source code represents 15% of violations, followed by intellectual property at 13% and passwords and API keys at 12%. This distribution highlights the continued importance of protecting compliance-sensitive and proprietary information, reinforcing the need for strong DLP controls and well-governed AI deployments.

Organizations in Europe are taking a cautious and risk-aware approach to AI adoption, with many choosing to block specific applications due to security, privacy, and compliance concerns. While policies vary by organization, certain tools are restricted more frequently than others, reflecting where perceived risk is highest. In regulated environments, blocking entire categories of AI applications can provide more consistent protection than managing individual tools.

Particular Audience is the most frequently blocked AI application at 44%, followed by ZeroGPT at 37% and DeepSeek at 36%. These applications often raise concerns around data handling transparency, personalization features, or limited visibility into how user data is processed and retained. These patterns indicate that organizations in Europe are not only reacting to risks posed by specific applications, but are also reinforcing broader governance strategies to ensure AI usage aligns with security and compliance requirements.

AI adoption continues to expand across organizations in Europe, with AI capabilities increasingly embedded into both operational and customer-facing workflows. Adoption now spans multiple layers: 64% of users are using AI applications directly, while an even larger share (95%) are using applications that incorporate AI-powered features indirectly. In addition, 89% of users are interacting with AI applications that rely on user data for training.

This widespread and multi-layered adoption highlights how deeply AI is becoming integrated into everyday workflows, often extending beyond explicit usage into embedded functionality within commonly used tools. It also reinforces the growing need for strong governance and data protection, as sensitive information may be exposed not only through direct interaction with AI applications but also through AI capabilities operating behind the scenes.