Core Viewpoint - The article discusses the challenges and potential of AI companies focusing on To G (government) projects, using Zhipu as a case study, and compares it to the previous AI "four dragons" that faced difficulties in profitability and sustainability [2][3]. Group 1: Zhipu's Business Model - Zhipu's revenue structure is heavily reliant on To G AI model privatization deployment, accounting for 84.8% of its total revenue, reminiscent of the previous AI four dragons [6]. - The company has faced significant losses, with cumulative losses exceeding 6.2 billion from 2022 to 2025, while total revenue was only 685 million [8]. - The challenges of To G projects include high delivery costs, low margins, long approval cycles, and slow payment processes, which have led to an accelerated increase in Zhipu's losses [8][10]. Group 2: Comparison with Previous AI Companies - The article highlights that Zhipu's business model mirrors that of the previous AI four dragons, which also struggled with profitability despite initial revenue spikes from government contracts [5][19]. - The previous AI four dragons faced a similar "profit dilemma," where high revenue did not translate into sustainable profits, leading to a retreat from the To G market [9][18]. - Unlike the previous companies, Zhipu has diversified its funding sources by attracting local capital, which may provide a more stable foundation for its operations [20]. Group 3: Market Dynamics and Future Prospects - The article notes that the capital market's perception of To G projects has shifted from enthusiasm to skepticism, raising questions about Zhipu's future performance [15][18]. - Zhipu's strategy of leveraging local partnerships and investments may offer a pathway to navigate the challenges of the To G market, potentially leading to a more sustainable business model [21][22]. - The company is encouraged to explore additional revenue streams beyond To G projects, such as consumer-oriented products, to enhance its market valuation and profitability [29].