This chapter investigates corporate strategies in the development of large language models (LLMs), such as ChatGPT, through the lens of law and economics. A key question is whether public announcements of LLM initiatives in Japan constitute “vaporware,” a practice of premature or misleading disclosure with potential antitrust implications. Using a stock price event study, the analysis evaluates market reactions to announcements by major Japanese firms. The results show no significant cumulative abnormal returns, indicating that investors did not perceive these announcements as deceptive or manipulative. Instead, the market responded with stability, suggesting that LLM development initiatives are assessed within a reasonably efficient and competitive environment. These findings imply that concerns about vaporware in this sector may be overstated, though the costs of regulation and industrial policy remain relevant. The study highlights the need for careful policy design to support innovation while maintaining fair competition in the rapidly evolving field of artificial intelligence.