New Zscaler AI Security Report Reveals an Over 3,000% Surge in Enterprise Use of AI/ML Tools
Zscaler has released its ThreatLabz 2025 AI Security Report, revealing a dramatic 3,000% year-over-year increase in enterprise AI/ML tool usage. The study, analyzing over 536 billion AI transactions from February to December 2024, shows ChatGPT leading with 45.2% of all AI/ML transactions, while also being the most-blocked application.
Key findings include enterprises blocking 59.9% of AI/ML transactions due to security concerns, with total data sent to AI tools reaching 3,624 TB. The Finance & Insurance sector leads AI adoption (28.4% of transactions), followed by Manufacturing (21.6%), Services (18.5%), Technology (10.1%), and Healthcare (9.6%). The top five countries generating AI/ML transactions are the US, India, UK, Germany, and Japan.
The report highlights emerging threats from agentic AI and DeepSeek, emphasizing the need for Zero Trust security approaches to combat AI-powered cyber threats.
Zscaler ha pubblicato il suo Rapporto sulla Sicurezza AI ThreatLabz 2025, rivelando un drammatico aumento del 3.000% anno su anno nell'uso degli strumenti AI/ML da parte delle imprese. Lo studio, che analizza oltre 536 miliardi di transazioni AI da febbraio a dicembre 2024, mostra ChatGPT in testa con il 45,2% di tutte le transazioni AI/ML, risultando anche l'applicazione più bloccata.
I risultati chiave includono il fatto che le imprese bloccano il 59,9% delle transazioni AI/ML a causa di preoccupazioni di sicurezza, con un totale di dati inviati agli strumenti AI che raggiunge i 3.624 TB. Il settore Finanza e Assicurazioni guida l'adozione dell'AI (28,4% delle transazioni), seguito da Manifattura (21,6%), Servizi (18,5%), Tecnologia (10,1%) e Sanità (9,6%). I primi cinque paesi che generano transazioni AI/ML sono gli Stati Uniti, l'India, il Regno Unito, la Germania e il Giappone.
Il rapporto evidenzia minacce emergenti provenienti da AI agentici e DeepSeek, sottolineando la necessità di approcci di sicurezza Zero Trust per combattere le minacce informatiche alimentate dall'AI.
Zscaler ha publicado su Informe de Seguridad AI ThreatLabz 2025, revelando un dramático aumento del 3,000% año tras año en el uso de herramientas AI/ML por parte de las empresas. El estudio, que analiza más de 536 mil millones de transacciones de AI desde febrero hasta diciembre de 2024, muestra a ChatGPT liderando con el 45.2% de todas las transacciones AI/ML, siendo también la aplicación más bloqueada.
Los hallazgos clave incluyen que las empresas bloquean el 59.9% de las transacciones AI/ML debido a preocupaciones de seguridad, con un total de datos enviados a herramientas de AI que alcanzan los 3,624 TB. El sector de Finanzas y Seguros lidera la adopción de AI (28.4% de las transacciones), seguido por Manufactura (21.6%), Servicios (18.5%), Tecnología (10.1%) y Salud (9.6%). Los cinco principales países que generan transacciones AI/ML son EE. UU., India, Reino Unido, Alemania y Japón.
El informe destaca amenazas emergentes de AI agentiva y DeepSeek, enfatizando la necesidad de enfoques de seguridad Zero Trust para combatir las amenazas cibernéticas impulsadas por AI.
Zscaler는 AI ThreatLabz 2025 보안 보고서를 발표하며, 기업의 AI/ML 도구 사용이 전년 대비 3,000% 증가했다고 밝혔습니다. 2024년 2월부터 12월까지 5360억 건 이상의 AI 거래를 분석한 이 연구는 ChatGPT가 모든 AI/ML 거래의 45.2%를 차지하여 가장 많이 차단된 애플리케이션이기도 하다고 보여줍니다.
주요 발견 사항으로는 기업들이 보안 문제로 인해 AI/ML 거래의 59.9%를 차단하고 있으며, AI 도구에 전송된 총 데이터는 3,624 TB에 달한다고 합니다. 금융 및 보험 부문이 AI 채택을 주도하며 (28.4%의 거래), 그 뒤를 제조업 (21.6%), 서비스 (18.5%), 기술 (10.1%), 의료 (9.6%)가 따릅니다. AI/ML 거래를 생성하는 상위 5개국은 미국, 인도, 영국, 독일, 일본입니다.
보고서는 에이전트 AI와 DeepSeek로부터의 새로운 위협을 강조하며, AI 기반 사이버 위협에 맞서기 위해 제로 트러스트 보안 접근 방식의 필요성을 강조합니다.
Zscaler a publié son Rapport de Sécurité AI ThreatLabz 2025, révélant une augmentation dramatique de 3 000 % d'une année sur l'autre de l'utilisation des outils AI/ML par les entreprises. L'étude, analysant plus de 536 milliards de transactions AI de février à décembre 2024, montre que ChatGPT est en tête avec 45,2 % de toutes les transactions AI/ML, tout en étant également l'application la plus bloquée.
Les principales conclusions incluent le fait que les entreprises bloquent 59,9 % des transactions AI/ML en raison de préoccupations de sécurité, avec un total de données envoyées aux outils AI atteignant 3 624 To. Le secteur Finances et Assurances est en tête de l'adoption de l'AI (28,4 % des transactions), suivi par la Fabrication (21,6 %), les Services (18,5 %), la Technologie (10,1 %) et la Santé (9,6 %). Les cinq pays générant le plus de transactions AI/ML sont les États-Unis, l'Inde, le Royaume-Uni, l'Allemagne et le Japon.
Le rapport met en lumière les menaces émergentes provenant de l'AI agentique et de DeepSeek, soulignant la nécessité d'approches de sécurité Zero Trust pour lutter contre les menaces cybernétiques alimentées par l'AI.
Zscaler hat seinen AI ThreatLabz Sicherheitsbericht 2025 veröffentlicht, der einen dramatischen Anstieg von 3.000% im Jahresvergleich bei der Nutzung von AI/ML-Tools in Unternehmen zeigt. Die Studie, die über 536 Milliarden AI-Transaktionen von Februar bis Dezember 2024 analysiert, zeigt ChatGPT als führend mit 45,2% aller AI/ML-Transaktionen, während es auch die am häufigsten blockierte Anwendung ist.
Wichtige Erkenntnisse umfassen, dass Unternehmen 59,9% der AI/ML-Transaktionen aufgrund von Sicherheitsbedenken blockieren, wobei die insgesamt an AI-Tools gesendeten Daten 3.624 TB erreichen. Der Finanz- und Versicherungssektor führt die AI-Adoption an (28,4% der Transaktionen), gefolgt von der Fertigung (21,6%), Dienstleistungen (18,5%), Technologie (10,1%) und Gesundheitswesen (9,6%). Die fünf Länder mit den meisten AI/ML-Transaktionen sind die USA, Indien, das Vereinigte Königreich, Deutschland und Japan.
Der Bericht hebt aufkommende Bedrohungen durch agentische AI und DeepSeek hervor und betont die Notwendigkeit von Zero Trust-Sicherheitsansätzen zur Bekämpfung von AI-gestützten Cyber-Bedrohungen.
- Massive 3,000% YoY growth in enterprise AI/ML tool adoption indicates strong market penetration
- Analysis of 536 billion AI transactions demonstrates substantial platform usage and data collection
- Strong market position in key industries with Finance (28.4%) and Manufacturing (21.6%) leading adoption
- Global presence with significant transaction volumes from major markets (US, India, UK, Germany, Japan)
- High blocking rate (59.9%) of AI/ML transactions suggests significant security concerns
- Emerging security threats from agentic AI and DeepSeek could impact adoption
- Potential risks from data leakage and compliance violations may affect enterprise implementation
Annual ThreatLabz Report Reveals 36x More AI/ML Transactions in World’s Largest Security Cloud, Stressing the Need for Zero Trust Everywhere to Enable Secure GenAI Adoption and Stop AI-Powered Threats
- ChatGPT is the most popular AI/ML application, accounting for nearly half of all AI/ML transactions (
45.2% ) and is also the most-blocked AI application, followed by Grammarly, and Microsoft Copilot as the second and third most-blocked applications, respectively - Agentic AI and open-source model DeepSeek are creating new opportunities for threat actors to weaponize AI and automate and scale their attack
- The top five countries generating the most AI/ML transactions are the United States, India, United Kingdom, Germany, and Japan
- The Finance & Insurance and Manufacturing industries generate the most AI/ML traffic, with
28.4% and21.6% share of all AI/ML transactions in the Zscaler cloud, respectively, followed by Services (18.5% ), Technology (10.1% ), Healthcare (9.6% ), and Government (4.2% )
SAN JOSE, Calif., March 20, 2025 (GLOBE NEWSWIRE) -- Zscaler, Inc. (NASDAQ: ZS), the leader in cloud security, today released the ThreatLabz 2025 AI Security Report, based on insights from more than 536 billion AI transactions processed between February 2024 to December 2024 in the Zscaler Zero Trust Exchange™platform, the largest in-line security cloud, which discovered real-world threat scenarios ranging from AI-enhanced phishing to fake AI platforms. This report also explores recent developments in areas that will undoubtedly influence AI in 2025 and beyond, including agentic AI, the emergence of DeepSeek, and the evolving regulatory landscape.
The report reveals a 3,000+% year-over-year growth in enterprise use of AI/ML tools, highlighting the rapid adoption of AI technologies across industries to unlock new levels of productivity, efficiency, and innovation. Enterprises are sending significant volumes of data to AI tools, totaling 3,624 TB, underscoring the extent to which these technologies are integrated into operations. However, this surge in adoption also brings heightened security concerns. Enterprises blocked
“As AI transforms industries, it also creates new and unforeseen security challenges,” said Deepen Desai, Chief Security Officer at Zscaler. “Data is the gold for AI innovation, but it must be handled securely. The Zscaler Zero Trust Exchange platform, powered by AI with over 500 trillion daily signals, provides real-time insights into threats, data, and access patterns—ensuring organizations can harness AI’s transformative capabilities while mitigating its risks. Zero Trust Everywhere is the key to staying ahead in the rapidly evolving threat landscape as cybercriminals look to leverage AI in scaling their attacks.”
Key Insights from the ThreatLabz 2025 AI Security Report
ChatGPT Dominates AI/ML Transactions, But Security Concerns Remain
ChatGPT emerged as the most widely used AI/ML application, driving
“We had no visibility into ChatGPT. Zscaler was our key solution initially to help us understand who was going to it and what they were uploading.”
—Jason Koler, CISO, Eaton Corporation | See the video case study
DeepSeek and Agentic AI: Innovation Meets Escalating Threats
AI is amplifying cyber risks, with usage of agentic AI and China’s open-source DeepSeek enabling threat actors to scale attacks. So far in 2025, we’ve seen DeepSeek challenge American giants like OpenAI, Anthropic, and Meta, disrupting AI development with strong performance, open access, and low costs. However, such advancements also introduce significant security risks.
Geographies Leading AI Adoption: US and India
The United States and India generated the highest AI/ML transaction volumes, representing the global shift toward AI-driven innovation. However, these changes aren’t occurring in a vacuum, and organizations in these and other geographies are grappling with increasing challenges like stringent compliance requirements, high implementation costs, and shortage of skilled talent.
Finance & Insurance Lead Enterprise AI Traffic by Industry
The Finance & Insurance sector accounted for
The Zscaler AI Advantage
Built on a true zero trust architecture, Zscaler delivers Zero Trust Everywhere, securing user, workload, IoT/OT communication using business policies, not network policies. Zscaler mitigates AI-powered threats by hiding applications and IP addresses from attackers, inspecting all traffic for threats, and ensuring users access only authorized applications—never full networks. This approach minimizes the attack surface, prevents lateral movement, and stops threats before they can cause harm. Zscaler protects its users against today's most sophisticated AI-driven threats by implementing the following:
- Zero Trust Foundation: Minimize the external attack surface through continuous verification and least-privilege access.
- Real-time AI Insights: Employ predictive and generative AI to deliver actionable insights that enhance security operations and digital performance.
- Data Classification: Leverage AI-driven classification to seamlessly detect and safeguard sensitive data across Zscaler’s Data Fabric.
- Threat Protection: Block AI-enhanced threats through continuous monitoring and response powered by the Zscaler Zero Trust Exchange.
- App Segmentation: Restrict lateral movement and reduce the internal attack surface with AI-driven, automatic app segmentation.
- Breach Prediction: Harness the power of Zscaler Breach Predictor that combines the power of generative AI and multi-dimensional predictive models.
- Cyber Risk Assessments: Leverages AI-generated security reports to continuously optimize your zero trust implementation.
Download the Full ThreatLabz 2025 AI Security Report
Download the full version of the 2025 AI Security Report here for more information about real-world threat scenarios, AI predictions, insights into AI regulations, and AI best practices.
Methodology
Analysis of 536.5 billion total AI and ML transactions in the Zscaler cloud from February 2024 to December 2024. The Zscaler global security cloud processes over 500 trillion daily signals and blocks 9 billion threats and policy violations per day, delivering over 250,000 daily security updates.
About ThreatLabz
ThreatLabz is the security research arm of Zscaler. This world-class team is responsible for hunting new threats and ensuring that the thousands of organizations using the global Zscaler platform are always protected. In addition to malware research and behavioral analysis, team members are involved in the research and development of new prototype modules for advanced threat protection on the Zscaler platform, and regularly conduct internal security audits to ensure that Zscaler products and infrastructure meet security compliance standards. ThreatLabz regularly publishes in-depth analyses of new and emerging threats on its portal, research.zscaler.com.
About Zscaler
Zscaler (NASDAQ: ZS) accelerates digital transformation so customers can be more agile, efficient, resilient, and secure. The Zscaler Zero Trust Exchange™ platform protects thousands of customers from cyberattacks and data loss by securely connecting users, devices, and applications in any location. Distributed across more than 150 data centers globally, the SASE-based Zero Trust Exchange™ is the world’s largest in-line cloud security platform.
Media Contact
Natalia Wodecki
press@zscaler.com
A photo accompanying this announcement is available at https://www.globenewswire.com/NewsRoom/AttachmentNg/9c2bf5d3-5720-4db8-bf1f-a9675f48840e
FAQ
What percentage of AI/ML transactions were blocked by enterprises according to Zscaler's 2025 report?
How much did enterprise use of AI/ML tools grow according to ZS's latest report?
Which industries are the top users of AI/ML according to Zscaler's 2025 report?
What is ChatGPT's share of AI/ML transactions in Zscaler's security cloud?
How much data was sent to AI tools through Zscaler's platform during the study period?
