New Innovations in Cyber Defense: Writuraj Sarma Leads the Charge with Agentic AI

Writuraj Sarma is considered a rising influencer in the field of cybersecurity and currently serves as Lead Product Manager for Samsung. He’s been at the forefront of creating agentic AI frameworks to help specifically counter the burgeoning cyberattack threat within e-commerce and retail industries. Sarma is an independent AI researcher with a deep expertise in…

Alexis Wang Avatar

By

New Innovations in Cyber Defense: Writuraj Sarma Leads the Charge with Agentic AI

Writuraj Sarma is considered a rising influencer in the field of cybersecurity and currently serves as Lead Product Manager for Samsung. He’s been at the forefront of creating agentic AI frameworks to help specifically counter the burgeoning cyberattack threat within e-commerce and retail industries. Sarma is an independent AI researcher with a deep expertise in defense systems, predictive analytics, and machine learning. His experience and expertise are what make him such an important voice today about necessary and beneficial conversations around cybersecurity.

His interest spiked after the big retail/e-commerce cyberattacks in May 2025. This experience compelled him to raise awareness about adopting agentic AI solutions. Sarma’s bottom line is that companies need to come out with limited pilot programs. Beyond this, they need to do better ROI (return on investment) vetting and avoid binding themselves to single vendor contracts. Yet his insights arrive at a moment when the need for more robust cybersecurity solutions has never been clearer.

The Rise of Agentic AI

Agentic AI is remaking the cybersecurity landscape. It moves the battlefield away from generally human-centered defensive measures to offensive and autonomous assets with the ability to learn and change the scenario. Sarma adds that these new inventions are becoming imperative as law enforcement has to keep up with the more advanced techniques that cybercriminals use these days.

“Agentic AI marks the shift from human-limited defense to autonomous resilience, always learning, always adapting, always protecting,” he stated. Collectively, this approach enhances our capacity to respond proactively to potential threats. It allows organizations to proactively identify and remediate vulnerabilities that malicious actors could exploit before they even have a chance.

Sarma’s analysis of the current landscape reveals three significant risk groups that organizations need to monitor closely: immature tools and hype, user-related challenges, and the potential leakage of personal information. Otherwise, he warns, too many companies will rush to deploy harmful agentic AI. Without a complete understanding of its capabilities or limitations, they endanger their future investments, and they’ll create strategies that miss the mark.

Lessons from the 2025 Cyberattacks

The combined May 2025 cyberattacks served as an urgent wake-up call for most businesses operating in the e-commerce and retail sectors. These attacks made it clear that strong security measures were necessary and that next-generation security solutions would be required to identify threats proactively. Sarma believes that if an agentic AI system had been operational during these attacks, it could have detected unusual login attempts—a crucial indicator of a potential breach.

“In the age of autonomous threats, hesitation is surrender,” Sarma warned, stressing the importance of vigilance and proactive defense strategies. His insights reflect a growing recognition that traditional cybersecurity measures may no longer suffice in an era where attacks are becoming more frequent and complex.

Agentic AI holds immense potential, but as Sarma cautions, there are risks. He estimates that by 2027 more than 40% of existing active projects in this innovative field may be left behind because of immature technology, undefined value proposition, or cost overrun. As organizations continue to pilot these solutions, they need to push past their novelty and learn what makes them work long-term (and what doesn’t).

The Future of Cybersecurity

Sarma’s vision for the future of cybersecurity rests on the complex and precarious integration of agentic AI technologies into existing infrastructures. He advocates for a reasonable middle ground. We need to give organizations the power and tools of autonomous systems, but pair them with rigorous governance to minimize risks of misuse and inherent bias.

“Agentic AI is our best defense—autonomous, predictive, and relentless against the evolving tide of cyberattacks,” he asserted. He’s passionate about using advanced technologies to stay competitive and that any business that doesn’t embrace them is doomed to fail. It’s no secret that in our ever-growing digital economy, companies need to innovate to survive.

Alexis Wang Avatar