Assessing the cybersecurity innovation frontier through measures of scientific output and technological inventions

This technical paper presents a novel, taxonomy-driven methodology for producing comparable indicators to track cybersecurity innovation across time, technologies and geographies. It uses data on scientific publications, which are treated as early signals of emerging cybersecurity knowledge, and patents, which capture the maturation of these ideas into technological solutions. A curated cybersecurity vocabulary validated by experts is identified and mapped to relevant Cooperative Patent Classification (CPC) classes to enable consistent measurement. This approach is complemented by well-established taxonomies of artificial intelligence (AI) and quantum computing technologies. Resulting indicators show that in 2024, cybersecurity patents are estimated to represent about 1.7% of all patents, a share similar to that observed in 2014 despite an increase in the overall number of patents. Over the same period, the share of cybersecurity-related scientific publications more than doubled, reaching 1% of all scientific publications in 2024, underscoring a significant increase in cybersecurity research.