Netskope 2026 Report Finds GenAI Data Policy Violations Doubled
Original report publisher: not verified. Report discovered via Awesome Annual Security Reports.
Summary
Netskope reports that genAI data policy violations doubled, averaging 223 incidents per organization monthly. Its 2026 report also highlights personal cloud app exposure, persistent phishing, and malware distribution through trusted services.
Key points
- The report draws on anonymized Netskope One usage data collected from October 1, 2024, through October 31, 2025.
- GenAI users tripled and prompt volume increased sixfold, while detected data policy violations doubled to an average of 223 per organization monthly; 50% of organizations lacked enforceable genAI data protection policies.
- Source code accounted for 42% of genAI data policy violations, followed by regulated data at 32% and intellectual property at 16%.
- Personal AI account use fell from 78% to 47% of AI users, but personal cloud app instances remained involved in 60% of insider threat incidents.
- Monthly phishing-link clicks declined 27%, from 119 to 87 per 10,000 users. Microsoft impersonation accounted for 52% of clicks on cloud phishing campaigns.
- Employee exposure to malware through GitHub was detected at 12% of organizations monthly, compared with 10% for OneDrive and 5.8% for Google Drive.
- Netskope recommends inspecting web and cloud downloads, blocking unnecessary apps, applying DLP to sensitive data transfers, and using least-privilege controls and continuous monitoring for AI agents.
Article Details
- Publisher
- Netskope
- Report Period
- 2024-10-01 to 2025-10-31
- Scope
- Enterprise genAI adoption, sensitive data exposure, personal cloud application use, phishing, malware delivery, and SaaS supply chain risks, based on anonymized Netskope One usage data.
- Sample Size
- Not disclosed
- Key Statistics
- Over the past year, SaaS genAI users tripled in the average organization, while monthly prompts increased sixfold from 3,000 to 18,000.
- Detected genAI data policy incidents doubled over the past year; the average organization recorded 223 incidents per month involving 3% of its genAI users.
- 50% of organizations lacked enforceable data protection policies for genAI applications, limiting visibility into actual sensitive data exposure.
- 60% of insider threat incidents involved personal cloud application instances.
- Monthly phishing-link clicks declined 27%, from 119 to 87 per 10,000 users over the past year.
- Recommendations
- Inspect all HTTP and HTTPS downloads across web and cloud traffic for malware.
- Block applications without a legitimate business purpose or with disproportionate risk; consider allowing approved applications and blocking others.
- Apply DLP policies to detect sensitive information sent to personal application instances, genAI applications, and other unauthorized destinations.
- Use Remote Browser Isolation when accessing higher-risk website categories, including newly observed and newly registered domains.
MITRE ATT&CK
T1189 · Drive-by CompromiseHidden iframe injections load malicious JavaScript and can trigger drive-by downloads.T1195.001 · Compromise Software Dependencies and Development ToolsThe report describes malicious npm packages attempting to distribute harmful code through developer workflows.T1199 · Trusted RelationshipAttackers abuse trusted SaaS integrations and API trust chains, including compromised Salesloft Drift connectors used for data theft.T1539 · Steal Web Session CookieReverse-proxy phishing kits steal session cookies in real time.T1566 · PhishingAttackers use phishing campaigns with counterfeit login pages and impersonated authentication or document-signing workflows to steal credentials.
Threat Actors
Vendors
NetskopeOrganization: NetskopeOpenAIAlready, organizations are seeing rapid adoption of agentic AI, both using SaaS services and AI platforms like Azure OpenAI.SalesforceIn one high-profile case, Salesforce detected suspicious API calls originating from non-allowlisted IP addresses via Gainsight integrations, prompting Salesforce to revoke associated access tokens, restrict
Products
Amazon BedrockCurrently, 33% of organizations use OpenAI services via Azure, 27% use Amazon Bedrock, and 10% leverage Google Vertex AI.Azure OpenAIAlready, organizations are seeing rapid adoption of agentic AI, both using SaaS services and AI platforms like Azure OpenAI.ChatGPT- **SaaS genAI use is rapidly increasing**: The number of people using SaaS genAI apps like ChatGPT and Gemini has increased threefold, while the number of prompts people are sending to the apps has increased sixfold inGitHubvictims through trusted channels, including software registries like npm and popular cloud apps like GitHub, OneDrive, and Google Drive.Google Drivechannels, including software registries like npm and popular cloud apps like GitHub, OneDrive, and Google Drive.Google GeminiInternal monitoring across organizations shows that a substantial share of employees are relying on tools such as ChatGPT, Google Gemini, and Copilot using credentials not associated with their organization.Microsoft 365 CopilotMicrosoft 365 Copilot reached 52% adoption, showing strong interest in AI features integrated into everyday workplace environments.Microsoft OneDrivethrough trusted channels, including software registries like npm and popular cloud apps like GitHub, OneDrive, and Google Drive.Netskope Onetraffic, to prevent malware from infiltrating your network. Netskope customers can configure their Netskope One NG-SWG with a threat protection policy that applies to downloads from all categories and applies toSalesloft DriftThe attackers abused unmonitored inter-app integrations, specifically Salesloft Drift, illustrating how compromised SaaS connectors can become an invisible conduit for data theft.Vertex AICurrently, 33% of organizations use OpenAI services via Azure, 27% use Amazon Bedrock, and 10% leverage Google Vertex AI.
Industries
BankingWhile cloud and SaaS applications remain frequent targets, banking portals now account for 23% of observed phishing lures, reflecting attackers' focus on financial fraud and account takeover.Government servicesGovernment services have also risen sharply to 21%, driven by attackers exploiting digital-ID and citizen-service portals for identity theft and tax fraud.