Kuaishou Technology is advancing plans to spin off its video generation business, KLing AI, paving the way for a potential initial public offering next year. The market has previously reported that KLing AI is seeking a funding round of $2 billion at a valuation of $20 billion, with discussions underway with investors including Tencent. On May 12, Kuaishou confirmed in an announcement that its board is evaluating a restructuring plan for KLing AI's assets and business, which may involve external financing. On that day, Kuaishou's Hong Kong stock opened sharply higher by about 10%. At the close, the stock price was HK$52.6 per share, up 1.94%, with a total market capitalization of approximately HK$228.7 billion. This development means that KLing AI, established less than two years ago, is targeting a valuation close to 70% of its parent company's total market cap, formally initiating the capitalization process for a leading player in China's AI video generation sector. The origins of KLing AI can be traced back to the internal technological expertise of Kuaishou's large model team. On June 6, 2024, Kuaishou officially launched its self-developed video generation large model, KLing AI. As the world's first publicly available tool for generating photorealistic, cinematic-quality videos, KLing AI initially enabled the generation of 2-minute, 30fps, 1080p high-definition videos, addressing industry pain points like unnatural motion and violations of physical laws in earlier models. Within just half a month of its launch, KLing AI introduced an image-to-video feature, opening a conversion channel from static to dynamic content for ordinary creators. On July 24, 2024, KLing AI fully opened its public beta, and the simultaneous release of a video continuation feature further met the demand for long-form video creation. In September 2024, Kuaishou collaborated with nine renowned directors, including Li Shaohong and Jia Zhangke, to launch the "KLing AI Director Co-creation Plan," achieving the first film-level short film entirely generated by AI. Entering 2025, KLing AI accelerated its technological iterations and organizational upgrades. In January 2025, KLing AI launched a multi-image reference mode, solving the challenge of multi-subject consistency in videos, and simultaneously released a standalone app. In April 2025, Kuaishou elevated the KLing AI division to a first-level department on par with e-commerce and commercialization, reporting directly to founder and CEO Cheng Yixiao, significantly boosting its strategic importance. Subsequently, KLing AI released multiple versions including 2.0, 2.1, and 2.5 Turbo. The 2.5 Turbo model, launched in September 2025, achieved comprehensive improvements in text comprehension, dynamic effects, and aesthetic quality, topping global authoritative AI benchmark tests in both image-to-video and text-to-video categories. By the end of 2025, KLing AI's global user base had surpassed 60 million, having generated over 600 million videos and providing API services to more than 30,000 enterprise clients and developers. Within a year and a half, KLing AI completed over 30 major product iterations, averaging a functional optimization every two weeks, making this high-frequency iteration capability one of its core strengths. KLing AI's rapid growth is clearly reflected in Kuaishou's 2025 financial results. The report shows Kuaishou's total revenue for 2025 grew 12.5% year-over-year to 142.8 billion yuan, with adjusted net profit at 20.6 billion yuan, up 16.5%. KLing AI contributed approximately 1.04 billion yuan in revenue for the full year, far exceeding the initial performance target of $60 million. Quarterly breakdown shows KLing AI revenue for the first three quarters of 2025 was 150 million yuan, 250 million yuan, and 300 million yuan respectively, reaching 340 million yuan in the fourth quarter, maintaining steady quarter-on-quarter growth. In December 2025, KLing AI's monthly revenue exceeded $20 million, corresponding to an annualized revenue run rate of $240 million. Entering 2026, its commercialization process accelerated further; as of January 2026, KLing AI's annualized revenue run rate had climbed to over $300 million, with monthly active users exceeding 12 million. Kuaishou indicated during an earnings call that KLing AI is expected to achieve revenue growth of over 100% year-over-year in 2026. KLing AI's commercialization primarily revolves around C-end memberships and B-end enterprise services. On the C-end, the continuous improvement of the membership system drives steady growth in paying users. On the B-end, its capabilities have widely permeated sectors including advertising and marketing, e-commerce, film and television, short dramas, animation, and gaming. Financial data shows that in Q4 2025, online marketing service spending driven by AIGC marketing materials reached 4 billion yuan, and generative recommendation large models and intelligent bidding models contributed approximately a 5% increase to domestic online marketing service revenue. To support KLing AI's sustained development, Kuaishou significantly increased related investments. Kuaishou's capital expenditure in 2025 was approximately 15 billion yuan and is projected to rise to 26 billion yuan in 2026. The additional 11 billion yuan in investment will mainly be used for computing power support for the KLing large model and other foundational large models, including server procurement and data center construction. However, KLing AI remains in an investment phase, and its profitability has not yet been fully realized. Currently, video generation models in the industry generally face challenges such as high computing power costs and limited commercialization models, and KLing AI is no exception. How to maintain technological leadership while further enhancing commercialization efficiency will be a core issue it needs to address after the spin-off and potential listing. KLing AI's spin-off plan is not an isolated event. In recent years, major domestic internet companies have been spinning off their star AI businesses for independent operation, seeking separate capitalization, forming a wave of AI asset spin-offs and listings. Shortly before news of KLing AI's spin-off emerged, Baidu's AI chip company, Kunlunxin, submitted a confidential listing application to the Hong Kong Stock Exchange on January 1, 2026, and formally initiated listing counseling for the Shanghai STAR Market on May 7, planning a dual listing in both A-share and H-share markets. Kunlunxin's predecessor was Baidu's internal Smart Chip and Architecture Department; after the spin-off, it will focus on AI computing chip R&D, with Baidu retaining a controlling stake. Earlier, there were also rumors of a spin-off and listing for Alibaba's semiconductor unit, T-Head. As a leading RISC-V chip design company in China, T-Head has deep expertise in the AIoT chip field; a spin-off would allow it to engage with capital markets more flexibly. Additionally, Tencent-backed AI startup StepFun recently completed a $2.5 billion funding round, dismantled its variable interest entity structure, and is accelerating its push for a Hong Kong IPO, potentially becoming "the third major domestic large model stock." Behind the trend of major tech firms spinning off AI businesses are multiple factors. First, the valuation logic for AI businesses differs from that of the parent company's traditional operations; a spin-off can lead to more reasonable market pricing. Taking KLing AI as an example, if valued at $20 billion, its valuation would be close to one-third of Kuaishou's current market capitalization, and a spin-off would significantly enhance Kuaishou's overall value. Second, independent operation grants AI businesses greater decision-making autonomy, helps attract top talent, and facilitates external collaborations. Finally, a spin-off opens independent financing channels for the AI business, alleviating the parent company's funding pressure, especially against the backdrop of the AI industry's need for sustained, high-intensity investment. For KLing AI, a spin-off and potential listing present both opportunities and challenges. On one hand, an independent listing would provide ample funding support for technology R&D and market expansion, while enabling more flexible collaboration across the industry chain. On the other hand, competition in China's AI video generation sector is intensifying, with models like ByteDance's Seedance 2.0 and Alibaba's HappyHorse相继推出. KLing AI needs to continuously maintain its technological edge to secure a firm footing in the fierce market competition. As more AI businesses head to the capital markets, competition in China's AI industry will enter a new phase. In the future, technological prowess, commercialization capability, and capital operation skills will become key determinants of success. Whether KLing AI can successfully list and become the "first AI video stock" warrants continued market attention.