Link: Japan goes light on AI regulation to court investment - Nikkei Asia

Japan is looking to lure AI investment away from Europe and elsewhere by becoming the "most AI-friendly country in the world," with a regulation-light approach. While the European Union recently passed the AI Act, billed as the world's first comprehensive law governing the use of artificial intelligence, the Japanese government's AI Strategic Council, which is leading the debate in Japan, is focusing on industry-led oversight. "None of the members is seriously calling for a 'hard law' to regulate AI," said a person familiar with the council's discussions. "We shouldn't prioritize regulation over innovation." "Japan is not thinking of implementing strict regulation, and there will be as few regulations as possible," said Masaaki Taira, chairman of the LDP's project team on AI. "We are asked that question from overseas tech companies quite often." He said Japan is pursuing a "slightly loose" AI policy compared with the EU in hopes of attracting investment away from Europe and elsewhere. "By making subtle changes, or angling the rules [to keep them business-friendly], it becomes easier to attract talent and money," he added. Japan's lack of globally competitive AI services has left it largely reliant on American tech companies like Microsoft-backed OpenAI for the advanced AI it needs to make its economy more efficient. This has also led to a "digital deficit," as much of Japan Inc.'s spending on software, data centers, underwater cables and semiconductors goes to North America, a drain exacerbated by the yen's recent weakness. But the road to the AI promised land may have some bumps. Japanese voters recently got a taste of what malicious AI can do, from scams using AI-generated messages to deepfake clips of a speech purportedly by Prime Minister Fumio Kishida. #

--

Yoooo, this is a quick note on a link that made me go, WTF? Find all past links here.