-- As enterprises grapple with the rapid integration of artificial intelligence into critical systems, a new line of thinking is emerging at the intersection of AI architecture and cybersecurity. Recent publications by Harsh Verma, Principal Software Engineer for AI at Palo Alto Networks, are contributing to a broader industry shift one that reframes how organizations think about trust, predictability, and security in AI systems.
Following his dual recognition at the 2026 Cybersecurity Excellence Awards, Verma’s latest analyses published through Forbes Technology Council and the RSA Conference platform are gaining traction among cybersecurity leaders and enterprise architects. The Cybersecurity Excellence Awards are presented by Cybersecurity Insiders, backed by a community of over 600,000 security professionals, and have recognised achievement across the cybersecurity community for more than a decade.
At the core of his work are two converging ideas: the rise of deterministic AI architectures and the growing threat of what he describes as the “clean attack problem.”
From Intelligent to Predictable: Redefining Enterprise AI Priorities
Verma's recognition reflects his work at the intersection of artificial intelligence, cybersecurity, and autonomous agent architectures, focusing on how modern enterprises can evolve toward real-time, self-directed security systems through agentic AI.
A significant dimension of his published work at RSAC addresses what he identifies as the clean attack problem: AI-assisted cyberattacks that operate entirely within legitimate workflows using valid credentials, approved API interactions, and authorised system access. In an analysis published on the RSA Conference platform, Verma argues that this development fundamentally disrupts the anomaly-based detection model on which cybersecurity has historically relied.
"Cybersecurity used to focus on detecting abnormal behavior," Verma writes. "AI systems are disrupting that model because attacks may now be carried out inside trusted workflows, with legitimate permissions, and normal operational patterns. Valid actions can produce malicious outcomes."
Toward Intent-Based Security in Autonomous Systems :
His proposed response is a shift toward intent-based security, a framework that evaluates not simply whether an action was authorised but why a sequence of actions is occurring, whether it aligns with the user's original request, and whether autonomous agents have drifted from their intended purpose in real time.
In a parallel analysis published in Forbes as a member of the Forbes Technology Council, Verma addresses how enterprises are evaluating AI systems more broadly, arguing that the next premium tier of AI will not be defined by capability but by reliability.
"The premium product in AI is no longer intelligence alone," Verma writes at forbes. "It is engineered trust. And trust begins with predictability, and predictability is where that engineering begins."
The Forbes analysis examines why hallucinations have moved from a technical concern to a business risk, and why large organisations are shifting from model-first to system-first thinking, with deterministic architectures built around retrieval-augmented generation, orchestrated workflows, and human approval checkpoints becoming the standard for enterprises in healthcare, financial services, and critical infrastructure.
Recognition Across the Industry: Community Validation and Industry Momentum
Verma’s recent publications build on a growing body of recognition and influence within the cybersecurity and AI communities. In 2026, he received both the AI Security Innovator of the Year award and the Community Choice Award at the Cybersecurity Excellence Awards, The 2026 awards add to a growing body of professional recognition. The Community Choice Award, the only Cybersecurity Excellence Award determined directly by community voting, reflects the visible engagement Verma has earned across his network and the wider cybersecurity community. The AI Security Innovator of the Year category attracted some of the strongest community support across the 2026 programme, reflecting what the awards organisers describe as a rapid shift in cybersecurity priorities toward AI-native defence architectures. The category itself is among several focused on AI that did not exist in the programme two seasons ago.
Harsh Verma’s work at Palo Alto Networks aligns closely with the company’s latest advancements in AI-driven network security, including the introduction of Strata Cloud Manager and Network Security Agents in PAN-OS 12.2. These innovations reflect a shift toward unified, AI-powered security operations where organizations can centrally manage complex environments, enforce Zero Trust policies in real time, and leverage intelligent agents to autonomously investigate, plan, and execute security workflows. By combining deep visibility, predictive analytics, and governed AI-driven automation, these systems are designed to scale security beyond human operational limits while maintaining control and reliability, reinforcing the transition from reactive defense models to proactive, intent-aware, and deterministic security architectures.
Advancing AI Knowledge: Google Developer Expert in Cloud AI
Harsh has also been recognised as a Google Developer Expert in Cloud AI, awarded to individuals who combine deep technical expertise with meaningful contributions to the global developer community. In this role, he shares knowledge on AI agents and agentic systems, helping developers and enterprises understand how to design, deploy, and scale autonomous AI solutions through talks, mentorship, and technical content.
Beyond his work at Palo Alto Networks, Verma serves as an advisor at Berkeley SkyDeck, collaborating with early-stage founders building in AI and enterprise security, helping startups bridge the gap between cutting-edge AI innovation and production-grade security requirements.
This combination of hands-on engineering, published thought leadership across Forbes and RSA Conference, community contribution as a Google Developer Expert, and ecosystem enablement through Berkeley SkyDeck positions Verma at a unique intersection of industry execution and broader field development.
About Harsh Verma
Harsh Verma is a Principal Software Engineer for AI at Palo Alto Networks, specializing in AI security architecture, agentic systems, and enterprise AI governance. He is a Senior Member of IEEE, a member of the Forbes Technology Council, and a Google Developer Expert in Cloud AI. His work focuses on building secure, scalable, and reliable AI systems for enterprise environments.
About the Cybersecurity Excellence Awards
The Cybersecurity Excellence Awards are an established global recognition programme honouring companies, products, and professionals advancing cybersecurity worldwide. Presented by Cybersecurity Insiders and backed by a community of over 600,000 security professionals. For more information visit cybersecurity-excellence-awards.com.
Contact Info:
Name: Gianmarco
Email: Send Email
Organization: Xraised
Website: http://xraised.com
Release ID: 89202999
Google
RSS