At GISEC 2025, one of the world’s largest cybersecurity gatherings held this week in Dubai, the dominant narrative was clear: artificial intelligence (AI in Cybersecurity ) now plays a paradoxical role in global cybersecurity. As emphasized by the event’s theme—“Securing an AI-Powered Future”—AI is both a formidable defense mechanism and a powerful weapon in the hands of cybercriminals. With over 25,000 visitors from 160 countries and more than 750 cybersecurity firms participating, the event served as a sobering reminder of the growing complexity of digital threats.
In his keynote address, Eugene Kaspersky, CEO and founder of Kaspersky, summed up the new reality: “We are living in a world where it’s very much AI vs. AI—a defense against attack.” He warned that while AI holds promise for safeguarding digital environments, it equally empowers cyber attackers with sophisticated tools. The industry, he stressed, must double down on efforts to ensure a secure future shaped by responsible AI development.
UAE Confronts Real-Time Threats and AI Risks
Mohamed Al-Kuwaiti, head of the UAE Cybersecurity Council, painted a concerning picture of the threat landscape in his remarks. The UAE AI in Cybersecurity alone experiences more than 200,000 cyberattacks daily, most targeting government institutions. As he highlighted, managing this risk requires more than advanced technology—it calls for strong AI governance, human-data integration strategies, and a focus on ethical considerations in digital security.
At the UAE Cybersecurity Council pavilion, attendees observed live cyber threat monitoring and demonstrations of potential vulnerabilities in everyday smart environments. Models depicted smart stadiums, warehouses, and transportation systems—each showing how minor security breaches could spark major disruptions. For example, manipulating robotic systems in a smart warehouse could cripple logistics, while tampering with access systems at a stadium could trigger public safety crises. Similarly, cloning RFID in transportation systems threatens the reliability of automated tolling.
Microsoft’s CVP for Security, Aarti Borkar, added that generative AI introduces fresh vulnerabilities by creating new attack surfaces—ranging from manipulated prompts and model outputs to orchestrated AI behaviors. This complexity adds another layer to the traditional threats faced by networks and data systems.
AI in Cybersecurity Infrastructure, Blind Spots, and Strategic Collaborations
Beyond consumer tech and government data, AI threats loom large over national and industrial infrastructure. Richard Robinson, CEO of U.S.-based cybersecurity firm Cynalytica, warned of a critical oversight in protecting what he calls “invisible layers” of industrial control systems—namely analog, serial, and Ethernet-based communications. These layers, often neglected in mainstream cybersecurity protocols, remain a major vulnerability.
To address this gap, Cynalytica announced a new partnership with global consulting giant PwC during GISEC 2025. The collaboration aims to expand deployment of Cynalytica’s threat detection tools across critical infrastructure clients worldwide. This initiative is set to strengthen compliance with cybersecurity regulations and boost forensic capabilities in complex industrial environments.
As cybersecurity experts at GISEC 2025 made clear, defending against AI-powered attacks will demand a multipronged approach—one that combines innovation, regulation, and global cooperation to stay one step ahead in a rapidly evolving digital battlefield.
