Kakao and AI Safety Research Institute Collaborate on AI Safety Evaluation Framework
Kakao partners with the AI Safety Research Institute to enhance AI safety standards.
Kakao partners with the AI Safety Research Institute to enhance AI safety standards.
Kakao has signed a memorandum of understanding (MOU) with the AI Safety Research Institute to develop a comprehensive framework for evaluating AI safety. This collaboration aims to create a systematic approach for identifying risks associated with AI technologies.
The partnership will involve joint safety assessments and the development of evaluation tools that meet global standards. Kakao will provide its AI models and agents for evaluation, while the AI Safety Research Institute will refine assessment methodologies.
The initiative will unfold in three phases, starting with safety assessments of language models, followed by multimodal models and AI agents. In the final phase, both organizations will work together to create efficient evaluation tools for AI safety verification.
Kim Se-ung, Kakao's AI Synergy Performance Leader, emphasized the importance of prioritizing safety alongside innovation, stating, "We will continue to strive to deliver AI models and services that prioritize safety and trust."
As AI technology evolves, the need for proactive risk assessment becomes increasingly critical. Kim Myung-ju, director of the AI Safety Research Institute, echoed this sentiment, highlighting their commitment to building a robust evaluation framework.
