Google Korea announced that on the 18th, it held the 3rd 'Responsible AI Forum' comprised of experts from various fields such as law, academia, IT industry, and startups, and discussed the direction of safe and beneficial AI utilization through explainable AI (XAI).
The 3rd forum was held in conjunction with the 'Safer with Google' event, a forum for discussion among experts to create a safe online environment. The forum mainly discussed ways to preemptively manage potential threats that arise from the advancement of AI technology, such as privacy invasion, cybersecurity threats, bias, and discrimination, and further secure the reliability and safety of AI through explainable AI.
Professor Jaesik Choi of KAIST Kim Jae-chul AI Graduate School and Director of KAIST XAI Research Center, an authority on explainable AI and the chairman of the 'Responsible AI Forum', introduced the operating principles of explainable AI and its use cases in various fields such as healthcare, finance, manufacturing, and government, while also talking about the future prospects and implications of explainable AI, such as AI regulatory trends, ethical issues, and expected effects that are actively discussed around the world.
Professor Choi said, "Explainable AI will significantly improve the transparency and reliability of AI technology by verifying the basis and algorithm for AI decision-making and explaining them in a way that humans can understand." He added, "As the world is focusing on securing AI safety, I hope that we can continue meaningful discussions for the development of AI at the 'Responsible AI Forum' in the future."
Google DeepMind, which has created AI models such as AlphaGo, also announced a toolkit for responsible AI use, such as the Language Interpretability Tool (LIT), and its future prospects.
“To enable more developers to implement responsible generative AI best practices, Google has developed the Responsible Generative AI Toolkit,” said Ludovic Ferran, product manager at Google DeepMind. “With the Responsible Generative AI Toolkit, developers can now standardize policy compliance, create content that respects diverse cultures and opinions, appropriately detect and control harmful content, provide sophisticated performance evaluation and risk measurement, and provide transparent feedback.”
In the discussion that followed the presentation that day, the 17 experts who attended as forum members discussed the following issues that have recently become hot topics: ▲the relationship between explainability and AI technological innovation, ▲the ethical and social impact of explainable AI, ▲ways to apply explainable AI in the legal system and policy domain, and ▲ways to simultaneously achieve explainable AI enhancement and AI technological innovation.
Shin Kyeong-ja, Google APAC Platform & Device Marketing and Google Korea Marketing Director, said, “Explainable AI, which has become more important with the emergence of generative AI, is one of the topics that is attracting attention in the development and innovation of responsible AI technology,” and “We hope that this forum will serve as an opportunity to continue more specific and in-depth discussions to secure the reliability and safety of AI while continuing innovation.”
Meanwhile, the Responsible AI Forum is an expert forum organized by Google Korea in February 2024 to seek responsible AI development and utilization methods through discussions on policy and regulatory issues and socio-ethical issues following the development of AI technology. Chaired by Professor Jae-sik Choi of KAIST, 14 experts from various fields such as law, policy, IT/technology, startups, and digital responsibility will participate as members of each forum.
The Digital Responsibility Committee was launched in early 2024 as part of Google Korea's efforts to create a responsible digital ecosystem in Korean society to hear the opinions of domestic experts in various fields. The committee will oversee and operate the 'App Ecosystem Forum', 'YouTube Open Forum', and 'Responsible AI Forum', which are expert forums for each of Google Korea's major business areas such as Google Play, YouTube, and Google AI, while activating exchanges between the forums to provide a constructive communication platform to lead the development of the Korean digital ecosystem. Each forum member consists of 39 experts who have been actively working in various fields for the development of the digital ecosystem, including IT/technology, law, media/content, business, administration, and consumer protection.
