In this bonus episode of season 5 of the Cyber Security District Podcast, we sit down with Marco Ramilli, an internationally recognised cybersecurity expert, researcher, TEDx speaker, and founder of IdentifAI, a company building advanced detection tools to verify the authenticity of digital content. With deepfakes rising across banking, insurance, media, and online identity verification, Marco’s work sits at the center of one of the most urgent challenges in cybersecurity: can we still trust what we see?
Marco’s career spans nearly twenty years across academia, malware research labs, large-scale incident response, and multiple companies, including a startup that was later integrated into Google’s VirusTotal. Today, he leads IdentifAI, where he is developing AI-powered tools to detect manipulated images, audio, and video with forensic precision.
Meet Marco Ramilli
Marco’s journey into cybersecurity began not in a formal training program but through experimentation and a moment that nearly got him expelled from university. He recalls attempting to test the security of an online exam platform, and accidentally triggering a major alarm. Instead of punishment, the university invited him to help solve the issue.
“That was when I realized cybersecurity was not only exciting,” he says, “but something I could dedicate my life to.”
His academic career led him to UC Davis, where he researched malware behavior and the security of digital voting systems. This exposure to real-world cyber risks laid the foundation for his future work in malware analysis, reverse engineering, and threat detection.
From Founder Struggles to a Major Acquisition
In 2014, Marco launched his first cybersecurity company in Italy. The early years were difficult: the team operated at break-even for more than three years, with the market not yet ready for advanced defensive analysis.
“We were surviving, not growing,” Marco explains. “But those years taught me resilience more than any success could.”
His persistence paid off. The company eventually grew to over 40 people and was later acquired and integrated with VirusTotal, Google’s cybersecurity platform. That experience shaped his understanding of how to build scalable security products, and inspired him to focus on the next big problem.
Why He Built IdentifAI: The Deepfake Explosion
Marco believes the shift to AI-generated content represents a fundamental turning point in cybersecurity.
“Content used to be deterministic. Now it’s probabilistic,” he explains. “You can no longer trust your eyes or your ears. We need tools that verify reality.”
IdentifAI uses a suite of AI and forensic techniques to detect manipulated content across images, videos, and audio. Instead of trying to identify the specific AI model used, the system analyzes artifacts left behind by the manipulation itself.
“We’re not detecting the tool. We’re detecting the manipulation,” Marco says. “That’s the only scalable way to do this as models evolve daily.”
The platform is currently used to detect insurance fraud, synthetic identities in banking, falsified scouting videos in sports, and manipulated media inside corporate environments.
“Fraudsters are already using AI. Organizations must catch up, or they’ll be blindsided.”
The Future of Security: From Deterministic to Probabilistic Thinking
Marco argues that cybersecurity is undergoing an irreversible shift.
“We are moving from deterministic security to probabilistic security,” he says. “The tools must change, and the mindset must change with them.”
As deepfakes, synthetic voices, and manipulated documents proliferate, organizations will need to rethink how they verify identity, authenticate content, and assess risk. Marco believes that deepfake detection will soon become a standard layer of digital trust, similar to spam filtering or endpoint protection today. His final message to CISOs and security leaders is clear:
“Stay curious. Curiosity is what keeps you ahead. It’s what will save us.”
Learn more about identifAI here.
Watch the full episode on YouTube
