Thales has unveiled a global cybersecurity response to the emerging risks posed by Frontier AI models, as increasingly capable AI systems introduce new possibilities for automated vulnerability discovery, exploit development and cyberattack execution.
Announced at the Thales Cyber Summit in Paris, France, the framework addresses the growing autonomy, reasoning capabilities and scale of Frontier AI systems. Thales said these capabilities are shifting AI from a technology that primarily assists human users towards systems capable of acting autonomously, creating new challenges for organisations, particularly those operating critical infrastructure.
“Frontier AI models represent a new inflection point for cybersecurity,” said Philippe Vallée, Executive Vice-President, Cybersecurity & Digital Identity, Thales. He said organisations need to understand their exposure, anticipate emerging risks and maintain resilience and rapid recovery capabilities if attacks succeed.
Thales' approach combines cybersecurity technologies with strategic and operational expertise, with the objective of helping organisations move from reactive security measures towards continuously tested cyber resilience.
The framework comprises five interconnected building blocks: application estate security; AI-augmented detection and response; a converged cyberdefence platform; continuous testing and risk evaluation; and protection and resilience of critical assets.
According to Thales, the value of the framework lies in connecting these capabilities so that threat intelligence, vulnerability information, exposure assessments, detection, remediation and validation of security controls can continuously inform one another.
The company also highlighted its partnerships with technology providers to strengthen the assessment and mitigation of AI-powered cyber threats. Through these collaborations, Thales aims to provide customers with access to advanced AI capabilities specifically designed for cybersecurity applications.
The initiative reflects the increasing focus among cybersecurity providers on preparing organisations for AI-enabled threats, while using AI itself to strengthen detection, assessment and response capabilities.
