A new global report by Palo Alto Networks reveals a growing gap between generative AI (GenAI) adoption and security readiness, with 10% of enterprise GenAI applications categorized as high-risk. The “State of Generative AI 2025” report, based on traffic from 7,051 organizations, highlights rising vulnerabilities tied to unsanctioned AI use and inadequate governance across sectors.
The report underscores India’s rapid GenAI uptake, with tools like Grammarly, Microsoft Power Apps, and Copilot among the top-used applications. At the same time, organizations face a spike in data loss prevention incidents and threats such as “Shadow AI,” where employees use GenAI tools without IT approval.
AI growth accelerates, but so do risks
Global GenAI traffic surged 890% in 2024 alone. Following the January 2025 launch of DeepSeek-R1, traffic from DeepSeek-based tools jumped by 1,800% in just two months. While AI is transforming enterprise productivity, it’s also expanding the threat surface, especially in critical infrastructure sectors such as manufacturing and technology.
Data loss incidents linked to GenAI more than doubled over the past year, now accounting for 14% of all recorded security incidents. The study also flags continued concerns over jailbreaking attacks—where users manipulate models into producing unsafe or malicious content.
India’s AI ambitions require stronger guardrails
With India earmarking ₹500 crore in its Union Budget 2025 for the Centre of Excellence in AI for education, its national AI roadmap is ambitious. However, Palo Alto Networks warns that without parallel cybersecurity investment, GenAI’s widespread deployment could compromise national security and public trust.
Swapna Bapat, VP & MD for India and SAARC at Palo Alto Networks, notes that many Indian firms are unaware of the full extent of GenAI already embedded in daily workflows. “The priority now isn’t whether to use these tools,” she said, “it’s how to secure them without slowing people down.”
Securing GenAI: Recommendations for enterprises
The report calls on enterprises to:
Establish visibility and control: Track GenAI app usage and enforce access policies
Prevent data leaks: Deploy real-time inspection and centralized policy enforcement
Defend against advanced threats: Implement Zero Trust architectures to combat AI-driven cyberattacks
As GenAI continues to evolve, organizations are urged to balance innovation with adaptive cybersecurity strategies that protect proprietary data, ensure compliance, and enable secure scalability.
