Dive Brief:
- According to a report published by the security firm Netskope on Tuesday, the erratic adoption of artificial intelligence (AI) by enterprises presents significant security vulnerabilities.
- Numerous employees are utilizing AI tools via personal accounts that lack adequate safeguards, thereby circumventing their organization’s IT governance, which enables cybercriminals to manipulate these platforms and breach corporate cybersecurity.
- Netskope’s report underscores this growing risk profile: “This combination of novel AI-driven threats and legacy security concerns defines the evolving threat landscape for 2026.”
Dive Insight:
Shadow AI, a term denoting unregulated AI usage within organizations, has long been recognized as a critical issue, yet continues to be a formidable challenge as companies endeavor to integrate AI into their operations.
Nearly half (47%) of individuals engaging with generative AI platforms are doing so through personal accounts that remain unsupervised by their employers, according to Netskope’s analysis encompassing cloud security data from October 2024 to October 2025.
This unmonitored AI usage generates vulnerabilities that hackers could potentially capitalize on.
Netskope noted, “A substantial share of employees are relying on tools such as ChatGPT, Google Gemini, and Copilot, using credentials not tied to their organization.”
The data presents a nuanced view of personal AI usage trends. While the proportion of individuals employing personal AI applications has significantly decreased from 78% to 47% over the previous year, the utilization of enterprise-approved accounts surged from 25% to 62%.
Conversely, the number of users oscillating between personal and corporate accounts crept up from 4% to 9%, suggesting that companies still have considerable work to do in enhancing user convenience and feature offerings.
Engaging with personal AI tools in professional settings poses various risks, including potential breaches of regulatory compliance and unsecured API links between external AI services and internal company servers.
Data exposure emerges as one of the most prevalent ramifications of unregulated AI usage, with Netskope reporting a year-over-year surge in the “number of incidents of users sending sensitive data to AI apps,” averaging 223 such incidents monthly per organization.
Security specialists advocate that organizations enhance their defenses against shadow AI practices by prioritizing the establishment of robust AI governance frameworks.
Netskope commented, “The shift toward managed [AI] accounts is encouraging,” while also emphasizing that employee behavior can evolve at a pace that outstrips governance efforts.
The firm recommends that organizations enact “clearer policies, better provisioning, and ongoing visibility into how AI tools are actually being utilized across the workforce.”
Source link: Cybersecuritydive.com.
