Invited by NORDIS, on the 7th of October 2025 Marco Fontani, forensics director at Amped Software, will give an online talk on media forensics and the trustworthiness of digital visual content.
As synthetic media becomes increasingly sophisticated, the trustworthiness of digital visual content is severely threatened. The common reliance on automated “deepfake detectors” presents a significant risk to credibility in forensic and journalism. These tools often function as inscrutable “black boxes,” providing a verdict without the explainability necessary for rigorous verification, and they remain susceptible to adversarial manipulation.
This presentation argues for a necessary shift from simple automated detection to a comprehensive forensic framework. At the heart of this approach is the crucial distinction between two foundational but often conflated concepts. We define integrity as the state of a media file being unaltered after its creation, and authenticity as the accurate representation of an event that actually occurred. Understanding this difference is key, as media can be authentic but lack integrity (e.g., a photo shared via social media), or have perfect integrity while being completely inauthentic (e.g., a staged video).
The proposed framework provides a pragmatic methodology for building a defensible case for or against a piece of visual media. This multi-faceted approach complements the classic journalistic investigation with a forensic examination aimed at revealing inconsistencies either at the signal or physical level (e.g., in light and shadows).
The ultimate goal is to equip practitioners with a critical methodology that moves beyond a simple “real or fake” determination toward a more nuanced and robust assessment of any visual asset.
About Marco Fontani
Marco Fontani is Forensics Director at Amped Software, a software company developing image and video forensic solutions for law enforcement agencies worldwide.
He earned his MSc in Computer Engineering in 2010 and his Ph.D. in Information Engineering in 2014. His research focused on image watermarking and multimedia forensics. His PhD dissertation, “Digital Forensic Techniques for Splicing Detection in Multimedia Contents”, was awarded the Best 2015 PhD Thesis on Signal Processing by the Italian National Telecommunications and Information Theory Group (GTTI).
He participated in several research projects funded by the EU and EOARD, and authored/co-authored over 30 journal and conference proceedings papers. He has experience delivering training to law enforcement and has provided expert witness testimony on several forensic cases involving digital images and videos. He is a former member of the IEEE Information Forensics and Security Technical Committee and actively contributed to developing ENFSI’s Best Practice Manual for Image Authentication.
Participation can happen online using the following link: meet.google.com/mhv-echx-zcc