This research examines the concept of ‘locality of broadcasting’ as one of the principles underlying the mass media concentration exclusion principle, and proposes an evaluation method tailored to the actual states of local television stations (local stations). This study analysed employee profiles posted on the new graduate recruitment web pages of local stations using M-GTA to visualise how employees perceive their company’s locality and incorporate it into their specific job duties. Thus, this study proposes a ‘two-step evaluation of locality’ framework grounded in the composition of managements’ and employees’ places of origin and the disclosure of such information. This framework enables flexible evaluation of locality that reflects the actual states of local stations, and also has the potential to activate communication among diverse stakeholders centred on local stations. The findings of this study have shed light on the qualitative expansion of locality that could not be recognised through traditional external analyses, and have presented an effective and low-cost evaluation method to support the sustainable development of local broadcasters.

Vocaloid songs have come to occupy a significant position in Japanese popular music. Previous literature has frequently focused on the nonhumanness of Vocaloid songs, that is, they tend to make sense only when sung by nonhuman virtual singers and tend to be too technically demanding for humans to perform. This paper analyzes 487 Vocaloid hit songs on Niconico Douga from 2007 to 2023, employing topic modeling to examine chronological changes in lyrics and characteristics of themes preferred in derivative creations. Key findings include: there is a decline in songs emphasizing the potential of non-human singing voices or Vocaloid characters, while songs that are not based on the unique characteristics of Vocaloid have become more prominent in Niconico’s recent hit charts. In addition, songs that can be enjoyed without requiring full contextual interpretation, songs that allow for open interpretations across diverse contexts, and songs that facilitate memeification through formulaic or dialogic structures tend to be consumed more actively in derivative creative practices. These findings suggest that since the mid-2010s, Vocaloid hit songs on Niconico have increasingly extended beyond the specific framing of non-humanity and have diversified and generalized within a more open cultural environment. This tendency aligns with the result that songs which can be flexibly recontextualized without strong dependence on their original context continue to appear within the scene and are more likely to foster derivative works.

The rapid diffusion of AI simultaneously exacerbates traditional market failures─information asymmetry, externalities, oligopoly, and economies of scale─while creating novel institutional challenges arising from its technical features, such as risk uncontrollability and limited explainability. Governments respond through two domains: governance for risk management and industrial policy for competitiveness. These domains, however, are mutually reinforcing rather than separate: governance builds trust that stimulates markets, while industrial policy provides the technological foundation for effective governance. This paper, grounded in theories of market failure, compares the latest institutional designs in the United States, Europe, China, and Japan. It highlights how the interplay between governance and industrial policy forms a virtuous cycle, and discusses theoretical and policy implications for the integrated design of national AI strategies.