Thales Unveils New Metamodel for Deepfake Detection

Thales Unveils New Metamodel for Deepfake Detection

In a ground-breaking development, the Thales team at cortAIx, the Group’s AI accelerator, has unveiled a new metamodel for deepfake detection. This revolutionary technology is a significant step forward in addressing the growing threat of deepfake content and ensuring the integrity of AI-enabled systems, particularly in mission-critical applications.

The Thales team, dubbed the Friendly Hackers, has created a powerful toolbox called BattleBox, designed to assess the robustness of AI models against attacks exploiting their vulnerabilities. These attacks include adversarial attacks and attempts to extract sensitive information. To combat these threats, the team has developed advanced countermeasures such as unlearning, federated learning, model watermarking, and model hardening.

The importance of deepfake detection cannot be overstated. Deepfakes are manipulated media, often in the form of videos or images, that convincingly depict individuals saying or doing things they never did. This technology has significant implications for various sectors, including politics, entertainment, and cybersecurity. Thales’s metamodel provides a crucial defense against the spread of misinformation and the potential manipulation of critical systems.

The Thales team’s expertise in this area was previously demonstrated in the CAID challenge, organized by the French defense procurement agency (DGA) in 2023. The challenge involved locating AI training data even after it had been deleted from the system to protect confidentiality. Thales’s ability to overcome this hurdle further solidified their reputation as leaders in the field of AI and cybersecurity.

Thales is renowned as a global leader in advanced technologies, specializing in defense and security, aeronautics and space, and cybersecurity and digital identity. With over 600 AI researchers and engineers, including 150 based at the Saclay research and technology cluster south of Paris, Thales is at the forefront of innovation in these domains.

The Group’s commitment to research and development is evident in its annual investment of close to €4 billion. Thales focuses on key innovation areas such as AI, cybersecurity, quantum technologies, cloud technologies, and 6G. With 81,000 employees in 68 countries and sales of €18.4 billion in 2023, Thales has established itself as a global leader in cutting-edge technologies.

The Thales team’s deepfake detection metamodel is a significant milestone in the ongoing fight against the proliferation of deepfake content. With its advanced countermeasures and robust toolbox, BattleBox, Thales is empowering organizations to protect themselves against the threats posed by deepfakes. This development marks a crucial step forward in ensuring the integrity and security of AI-enabled systems, ultimately making the world safer, greener, and more inclusive.

In a rapidly evolving technological landscape, the importance of staying one step ahead of potential threats cannot be overstated. Thales’s deepfake detection metamodel, developed by their talented team of AI experts, is a testament to their dedication to innovation and their commitment to securing critical systems. The future of AI and cybersecurity looks promising, thanks to pioneering efforts like those of the Thales team.


Written By

Jiri Bílek

In the vast realm of AI and U.N. directives, Jiri crafts tales that bridge tech divides. With every word, he champions a world where machines serve all, harmoniously.