The Japanese government has recently initiated a review of the artificial intelligence service Grok, owned by Elon Musk, following allegations that it can easily generate and disseminate sexualized images without the subjects' consent. This move places Japan within the expanding global coalition of regulators increasing pressure on social platform X and its AI functionalities. Kimi Onoda, the Minister for Economic Security who also oversees AI strategy, stated that the Cabinet Office has demanded that platform X strengthen its protective mechanisms to curb Grok's ability to generate sexually explicit deepfakes. The government has also submitted a written inquiry to X, seeking a detailed explanation of its specific measures to prevent the creation of deepfake images that violate personal privacy, intellectual property, and portrait rights.
This month, Grok faced user backlash and governmental scrutiny from countries ranging from Malaysia to Italy after it was revealed that users could easily exploit its features to alter photos into sexualized or derogatory content. Subsequently, the service has imposed partial restrictions on its image-generation capabilities and made them exclusive to paying subscribers. "If the situation does not improve, all options, including legal measures, will be considered," Onoda emphasized during a regular press conference on Friday, adding, "We will also take action if similar issues arise on other platforms."
This statement from a regulatory official in one of X's largest overseas markets follows investigative actions by regulators in numerous countries worldwide. Authorities in Canada, California, the European Union and member states like France are investigating whether images generated by Grok infringe upon human rights, while Malaysia, Indonesia, and the Philippines have already restricted access to the service within their borders. Earlier this week, xAI, the company responsible for developing Grok, announced it would disable the chatbot's feature for generating sexualized images of real people.
Onoda pointed out that "the fact that users can still generate such images is problematic," hinting that Japan may demand further action from the developer. However, she also stressed that the problem lies not with AI itself, underscoring the need for the government to explore diverse response strategies: "It's like having a knife placed in front of you; whether it's used for cooking or harming someone depends on the user."
Japan is seeking to strike a balance between AI regulation and development: on one hand, there is an urgent need to establish guardrails, while on the other, it strives to catch up with China and the United States in a field deemed crucial for national strategy. The comprehensive Japanese AI law, which took full effect this past September, does not include penalties but only authorizes the government to conduct investigations and issue administrative guidance in cases of violations. Policy discussions are currently focused on user education and technical-ethical solutions such as labeling for AI-generated image alterations.
The Cabinet Office is coordinating with the Ministry of Justice, the National Police Agency, and the Ministry of Internal Affairs and Communications to discuss potential measures, reflecting Japan's strategic balancing act between incentivizing AI innovation and managing its associated risks.
Comments