Perplexity launches Secure Intelligence Institute to advance AI security, privacy, and safety research

Secure Intelligence Institute

• Perplexity opens Secure Intelligence Institute for AI safety
• SII to focus on AI security, privacy, and trustworthy deployment
• Dr. Ninghui Li named inaugural director of the institute

Did our AI summary help?

Perplexity has announced the launch of the Secure Intelligence Institute (SII), a new initiative focused on advancing research in AI security, privacy, and safety. The institute aims to address emerging risks associated with increasingly capable artificial intelligence systems, including general-purpose agents and multimodal models.

Mission and focus areas

The company said that frontier intelligence systems are becoming more capable of reasoning, using tools, and executing complex tasks in real-world environments. As these systems evolve, they introduce new challenges related to security, data privacy, and safe deployment.

The Secure Intelligence Institute will conduct interdisciplinary research to study these challenges and develop practical defense mechanisms. Its work will focus on improving the trustworthiness of AI systems while contributing to broader ecosystem-wide safety efforts.

Perplexity added that the institute will combine internal research with collaborations across academia to expand the scope and impact of its findings.

Research and collaborations

The institute will support research across domains such as cryptography, privacy, robust machine learning, and trustworthy AI. It also plans to work with academic groups and publish research that contributes to real-world AI security practices.

Among its early efforts is BrowseSafe, an open-source benchmark and detection model designed to identify harmful content and threats across AI-driven web environments. The institute is also engaging with policy frameworks, including contributions to discussions around securing autonomous AI agents.

Leadership and industry context

Dr. Ninghui Li has been appointed as the inaugural director of the institute. He brings experience in security and privacy research, along with leadership roles in academic and industry bodies.

Perplexity said the initiative builds on its experience operating AI systems used by millions of users and enterprises. The company aims to translate research insights into practical improvements across its products while contributing to safer deployment of advanced AI technologies.