Kakao is partnering with the AI Safety Research Institute to create a safe and trustworthy artificial intelligence (AI) ecosystem.
On September 28, Kakao announced that it signed a strategic memorandum of understanding (MOU) with the AI Safety Research Institute to jointly evaluate AI safety and establish an evaluation system.
Last year, Kakao and the AI Safety Research Institute conducted a joint safety evaluation of its language model, 'Kanana-1.5-9.8B.' As AI technology rapidly advances, the importance of identifying and verifying potential risks in models has grown. Kakao confirmed a high level of safety compared to similar overseas models through this evaluation.
The MOU aims to expand collaboration and establish an objective and systematic safety evaluation framework for AI models and agents.
Under the agreement, Kakao will provide its AI models and agents for safety evaluation and support the infrastructure needed for evaluation and tool development. The AI Safety Research Institute will be responsible for conducting evaluations and enhancing methodologies. Both parties will jointly review evaluation results and negotiate the scope of public disclosure while collaborating on the development of AI safety evaluation tools.
The collaboration will be implemented in three phases. Initially, safety evaluations will be conducted before and after the deployment of language models, followed by an expansion of the evaluation scope to multimodal language models and AI agents. Finally, they plan to jointly develop evaluation tools that can more efficiently verify the safety of AI models and agents.
Through this agreement, Kakao aims to establish a system for proactively identifying and verifying risks in AI models while continuously enhancing its AI safety evaluation capabilities. The company also intends to contribute to creating a reliable AI ecosystem by securing global-level AI safety based on collaboration with domestic AI safety research institutions.
Kim Se-woong, Kakao's AI Synergy Performance Leader, stated, "This collaboration will enable Kakao's AI models to have a more objective and rigorous safety verification system. We will continue to strive to provide AI models and services that prioritize safety and trust alongside technological innovation."
Kim Myung-joo, Director of the AI Safety Research Institute, remarked, "As AI technology advances, it is crucial to have a system that proactively identifies and evaluates potential risks. We will do our best to establish an AI safety evaluation system that meets global standards through our collaboration with Kakao."
Kakao has established its own AI risk management system, the 'Kakao AI Safety Initiative (Kakao ASI),' to manage risks that may arise throughout the development, deployment, and operation of AI models by 2024. Based on this, the company is inspecting risk factors in its internal AI models and publicly disclosing the results of safety evaluations.
* This article has been translated by AI.
Copyright ⓒ Aju Press All rights reserved.