Bilibili's AI Business Moves From Cost Saving to Rebuilding the Content Platform

Deep News
Yesterday

Although the National Day holiday has only just ended, two shifts at Bilibili Inc. (NASDAQ: BILI) are already becoming clearer.

On one side, the content ecosystem continues to expand. In early October, the 2026 Bilibili Shanghai International Figure and Model Culture Expo, hosted by Bilibili, took place, and the connection among offline IP, creators, and users is still extending. On the other side, AI is rapidly entering Bilibili's content production system. On September 5, Bilibili announced the winners of its first AI Creation Open Competition; the event, which ran from June to August, ultimately generated more than 28,000 submissions, with non-professional developers such as students and full-time mothers accounting for more than 70%. Data disclosed in the second-quarter financial report shows that consumption time on AI knowledge content on Bilibili rose 72% year on year, while AI advertising revenue grew more than 100% year on year.

At the same time, Bilibili's second-quarter revenue was 7.94 billion yuan, up 8% year on year, adjusted net profit was 700 million yuan, and gross margin reached 37.2%; advertising revenue was 3.13 billion yuan, up 28% year on year, surpassing value-added services for the first time to become the largest revenue source, while gaming revenue fell 14% year on year. More than a month has passed since the second-quarter report was disclosed, and what really needs to be studied now is what happened after the report.

AI is turning from one of Bilibili's technical capabilities into part of its content ecosystem, and advertising happens to be the company's fastest-growing business. Management also made clear on the earnings call that Bilibili's AI investment is highly concentrated on video understanding, distribution, and creation. This gives Bilibili's AI story a more concrete analytical framework: how much new content AI can bring to a content platform is not the most important question; what matters more is whether it can lower content production costs, improve recommendation and search efficiency, and at the same time help advertisers complete content production and placement at lower cost. If this efficiency loop gradually works, Bilibili's future profit growth will have a path more worth studying than simply adding advertising slots.

According to statistics, Bilibili's advertising business has maintained growth of more than 20% for 14 consecutive quarters. Second-quarter advertising revenue of 3.13 billion yuan surpassed value-added services and became the company's largest single revenue source. First-half advertising revenue reached 5.72 billion yuan, up 29% year on year. Such growth is not simply the result of a recovery in the advertising market; more importantly, Bilibili has begun to raise the commercial value of each unit of traffic. Second-quarter advertising revenue growth came from industries including gaming, digital home appliances, online services, e-commerce, and automobiles; advertising for automobiles and footwear and apparel grew more than 60% year on year, while livestream commerce advertising in the home and home furnishings sector grew 59% year on year during the 618 period. At the same time, AI-related advertising revenue grew more than 100% year on year.

Competition in content platform advertising is entering a deeper layer. Previously, advertising platforms were mainly responsible for traffic distribution, and advertisers had to complete creative production, creative testing, and placement optimization on their own. After AI entered the picture, advertising platforms began to participate in content production itself and then use user data and algorithmic feedback to determine which creatives are more effective, gradually linking content production and advertising placement. TikTok has already been advancing a similar path, with Smart+ covering creative, audience, and budget management, while Symphony provides AI content generation capabilities; YouTube is also using tools such as AI dubbing to reduce the cost of cross-language distribution for creators. Bilibili does not need to copy the product forms of these two platforms, but the business problem is highly consistent: how to reduce content production costs while improving advertising conversion efficiency. Advertising scale growth has already been verified by the financial report, and what really needs to be observed next is whether the efficiency gains brought by AI can continue to be reflected in profit margins. For Bilibili, advertising becoming the largest revenue source is only a change in revenue structure; whether the advertising business can generate stronger operating leverage will truly determine the quality of this round of commercialization upgrading.

At present, Bilibili's largest long-term asset is still its UP host ecosystem, and content production has long faced a real constraint: the more vertical and professional the content, the more it tends to depend on creators investing time and professional ability. AI is changing part of that cost. Subtitles, dubbing, translation, editing, scripts, and material processing are all relatively standardized production links that used to require multi-person collaboration but can now be compressed into the hands of one creator by AI tools. Bilibili does not need to bear all production costs, but it can improve the efficiency of content supply through tools.

This year's AI Creation Open Competition provides a very intuitive sample. The event ran from June 5 to August 20, ultimately attracting more than 30,000 submissions, with non-professional developers accounting for more than 60%. The competition required participants to use AI to create runnable and interactive products and to publish the creation process on Bilibili. What is truly valuable about this is that it pushes AI from "watching content" to "producing content." In this way, users can both consume AI content and use AI to complete creations, while Bilibili provides a community environment for display, feedback, and dissemination. This forms a natural combination with Bilibili's own content ecosystem. In the second quarter, the number of UP hosts with more than 1,000 followers rose 30% year on year, those with more than 10,000 followers rose 21% year on year, and average daily video submissions rose 28% year on year. At the same time, consumption time on AI knowledge content increased 72%. Increased content supply and the spread of AI tools are interacting, but what truly determines commercial value is still content quality. After AI lowers the creation threshold, low-quality content, copyright, and community experience will all become new governance costs. What Bilibili needs to do is not simply increase AI content, but more importantly use recommendation, search, and community rules to turn more AI generation capability into effective content.

This is also the difference between Bilibili and pure AI tool companies. Bilibili already has a mature content distribution system, creator network, and community relationships. If AI can enter the three links of creation, understanding, and distribution, the value brought by AI will exceed the revenue of a tool and begin to show up more as efficiency improvements across the entire content system. In the second quarter, Bilibili's business divergence was already very obvious: advertising revenue was 3.13 billion yuan, up 28% year on year; value-added services were 2.97 billion yuan, up 5% year on year; gaming revenue was 1.39 billion yuan, down 14% year on year. Gross margin reached 37.2%, improving for 16 consecutive quarters. Based on the above data, the company's fundamental main line is gradually becoming clear: advertising is currently the strongest-growing business, gaming is still waiting for a new product cycle to improve, and profit margin has become an important indicator for judging the company's operating quality.

AI is participating in more and more links. On the advertising side, AI is being used for material generation, user matching, and conversion efficiency improvement; on the content side, AI lowers creators' production costs and expands supply; on the search side, the company has disclosed that search revenue doubled year on year; on the user side, consumption of AI knowledge content is increasing rapidly. If these changes continue, Bilibili's future revenue growth will not have to rely entirely on user scale. In the second quarter, Bilibili's average daily active users reached 117 million, average daily usage time was 113 minutes, and total user usage time rose 14% year on year. The fundamental variables truly worth tracking next are: whether the gaming business can return to growth with a new product cycle, which determines how much elasticity remains on the revenue side; whether the advertising business can maintain rapid growth, which determines how far the commercialization main line can go; and whether AI can continue to reduce content and advertising production costs, which determines whether profit margin improvement can continue.

From this perspective, Bilibili is entering a stage more worth studying. Previously, the market priced the company more around user growth, game cycles, and the content community. Now that advertising has become the largest revenue source, the importance of operating efficiency has clearly risen. AI's further entry into content and commercialization also gives Bilibili, for the first time, a long-term variable through which profitability can be observed from the production efficiency side. This path of course still needs to be delivered in results. AI content quality, copyright governance, community experience, and the performance of new games may all create disturbances, and advertising growth also needs to continue to be verified by quarterly financial reports. But so far, the company has given relatively continuous operating signals: user scale continues to grow, advertising continues to grow rapidly, gross margin keeps improving, AI advertising revenue is rising quickly, search commercialization is beginning to scale, and the creator ecosystem is also actively embracing AI.

What is truly worth watching about Bilibili is whether AI can enter the entire chain of content production, distribution, and commercialization. For the market, what needs to be tracked next is whether changes at Bilibili's product level can continue to translate into profit and cash flow. For Bilibili, the core value of AI lies in enabling the content, users, and creator resources it has accumulated to achieve higher production efficiency and commercial efficiency. If this path can continue, Bilibili will still be a content company, but the way the capital market looks at Bilibili may begin to change.

Disclaimer: Investing carries risk. This is not financial advice. The above content should not be regarded as an offer, recommendation, or solicitation on acquiring or disposing of any financial products, any associated discussions, comments, or posts by author or other users should not be considered as such either. It is solely for general information purpose only, which does not consider your own investment objectives, financial situations or needs. TTM assumes no responsibility or warranty for the accuracy and completeness of the information, investors should do their own research and may seek professional advice before investing.

Most Discussed

  1. 1
     
     
     
     
  2. 2
     
     
     
     
  3. 3
     
     
     
     
  4. 4
     
     
     
     
  5. 5
     
     
     
     
  6. 6
     
     
     
     
  7. 7
     
     
     
     
  8. 8
     
     
     
     
  9. 9
     
     
     
     
  10. 10