In our ongoing exploration of the STARLIGHT project's key contributors, we're delighted to introduce Dr Jorge García Castaño, who leads the Intelligence Security Video Analytics Department at Vicomtech. Vicomtech focuses on developing AI tools that sift through images and videos to find crucial evidence and predict cyberattacks using temporal information from large datasets.
At the heart of innovation
"In STARLIGHT, we're developing AI-based tools for identifying evidence in multimedia content and predicting cyberattacks." Furthermore, our work encapsulates Vicomtech's role in tackling some of STARLIGHT's most challenging technical aspects, including creating interfaces for AI tool benchmarking and organising ToolFests to demonstrate these technologies in action.
Tailored solutions through collaboration
Jorge and the Vicomtech team work closely with law enforcement agencies to understand their needs and deliver solutions that meet real-world challenges. This partnership is important in developing practical, impactful tools.
A step forward with AI
One innovative tool from Vicomtech can classify images from the dark web, dealing with complexities such as images belonging to multiple categories and learning new categories over time. Jorge explains that, "we need to apply incremental learning approaches," underscoring the tool's ability to adapt to new and emerging types of crime.
Ethical AI development
Vicomtech's commitment to ethical AI is relentless. "Our AI solutions aim to be explainable, fair, and respectful of human rights." This focus promotes the development of technologies with transparency and non-discrimination at their core, aligning with societal values and ethical standards.
To listen to Jorge on Vicomtech's contributions on secure and ethical AI applications, watch the full interview on our YouTube channel.