The article reports on the shift in generative AI use from individuals to organizational implementation, focusing on the latest developments, including a Notion survey, conditions in educational settings, moves toward legislative development, and the risks of shadow AI.
A Notion survey found that 64% of knowledge workers use AI at least 1 day a week. While individual use is progressing, organizational adoption is polarized, with 49% saying that the environment for use is “in place” and 51% saying it is “not in place.” The biggest barrier to expanding use within organizations is a lack of skills and training, cited by 33%. ベリサーブ has launched a game-based learning program for all of its approximately 2,000 employees, supporting the move toward implementation and operations through a 10-step design combining the use of ChatGPT Enterprise and Dify.
In education, a survey by Takeda Juku found that 84.7% of high school students use ChatGPT, while concerns about information accuracy have risen to 52.3%, along with concerns about dependence. A survey of parents also found that 54.3% viewed it positively, but nearly 35% struggle with how to let their children use it at home, again highlighting the importance of designing operational frameworks. The challenge in education is shifting away from “prohibition or acceptance” toward designing how and where information should be verified.
Legislative development has finally reached the starting line, with reports of moves to establish a legal framework addressing infringement of rights to a person’s likeness and voice. Voice actor Ogata Megumi (緒方恵美) commented that it was “finally the starting line.” As delays in制度 create risks, the development of governance is regarded as an urgent priority.
The main corporate risks are shadow AI and confidential information management. A Forbes JAPAN survey found that approximately 23% of managers enter confidential information into unauthorized AI tools, a high level that has been identified as a challenge requiring stronger information-systems functions and the introduction of continuous monitoring operations. As SaaS and AI integration advances, companies are moving from checks at the time of implementation toward ongoing operational monitoring.
The industrial impact is accelerating in multilingual applications and creative fields. Mavericks announced that NoLang would support a Korean-language UI, indicating an effort to reduce friction in the workplace. While the international expansion of AI-generated music continues to face rights-clearance challenges, Project Glasswing has released a model that accelerates vulnerability discovery and remediation.
In summary, the competitive focus of generative AI is shifting from “implementation” to “effective use, governance and organizational design.” Organizations that can integrate education, guidelines, rights clearance and information controls into a comprehensive framework, rather than focusing solely on usage rates, are entering an era in which they can differentiate themselves, and proactive operational design is identified as a potential source of future competitiveness.