Thank you for visiting the CTAI website.
There is growing enthusiasm surrounding the potential benefits AI can offer humanity. Alongside this optimism, however, there are growing concerns about harms arising from the use of AI, such as bias and discrimination, which could increase mistrust in AI, ultimately hindering the stable and sustainable growth of the technology and related industry. Bearing these challenges in mind, CTAI was established to lay the technical, ethical, and social groundwork necessary for trustworthy AI.
Bearing these challenges in mind, CTAI was established to lay the technical, ethical, and social groundwork necessary for trustworthy AI.
CTAI's objectives can be distilled into four main pillars.
Firstly,
we strive to define the concept of trustworthy AI (encompassing fairness, safety, explainability/transparency, robustness/security/privacy), establish evaluation criteria, and develop technical documents detailing standard procedures for verification and implementation.
Secondly,
we plan to assess whether AI models meet these criteria and conduct research to develop core technologies required to adapt models within feasible limits.
Thirdly,
we aim to create solutions that effectively implement these technologies.
Lastly,
we endeavor to design and promote educational programs that cultivate convergent talents
In pursuit of these research goals, CTAI unites researchers from diverse fields such as engineering, statistics, communication, law, economics and philosophy to engage in interdisciplinary and convergent research. We appreciate the National Research Foundation of Korea’s generous support for our research activities, and reiterate our commitment to materially contribute in establishing a social foundation for the trustworthy and secure use of AI.
Eun-Ju Lee
Director, Center for Trustworthy AI
Professor, Department of Communication
Seoul National University
