AI competition is transforming from models to applications, creating new opportunities for consumer and enterprise markets while restructuring industries
Leadleo Research Institute
As new AI model technologies continue to roll out, the focus of innovation is shifting from the model itself to the application. The model-driven application revolution is currently transforming business paradigms and user experiences across countless industries.
As AI capabilities extend from the cloud to real-world scenarios, the pressing question for all market participants is how to accurately capture user needs, accurately capture viable commercialization pathways, and harness new traffic patterns. The Leadleo Research Institute recently released the 2025 China Foundation Model Application Market Insight White Paper. It focuses on consumer and business-level applications and analyzes the market by examining core scenarios, traffic structures, differentiated demands and competitive strategies.
An application ecosystem is formed
The report notes that in the consumer market, large-scale model application layer products can be divided into three categories depending on the maturity of the business. It is an early investigation of more mature embedded applications, native AI applications development, and smart hardware. Embedded applications are the most mature, enhancing existing software for more efficient monetization. Native AI applications create whole new services around the model itself, but the business model is still being tested. Smart Hardware combines physical interaction with AI to not only have great potential, but also faces technical and adoption challenges.
Each type follows a different value implementation path. Embedded applications rely on user base for rapid monetization. Native AI needs to prove themselves through competition. Smart hardware needs to overcome technology, cost and user acceptance hurdles.
Globally, web-based AI applications for consumers have shown strong concentration effects. ChatGpt leads with around 4.7 billion visits per month, with Microsoft's new Bing following 1.53 billion people. Together they form the first layer, but Deepseek, Gemini, Prperxity, Chargetwe.ai, and Claude form the second layer, which is quite behind the leader.
According to the report, AI chat assistants and AI search engines are two dominant categories of web-based AI applications, accounting for more than 80% of traffic. This is because information search and interactive Q&A are the most frequent and basic online needs, and these applications can efficiently meet them. Their wide applicability and large user base, coupled with FirstMovers' advantages and integration with existing entry points, helped large products accumulate large users and validate their core values.
The difference between mobile apps and web-based applications is primarily in their interaction and functionality integration. Web applications do not require installation, platform-to-platform work, and are suitable for fast information access and text-based interaction. In contrast, mobile apps provide a smoother, more personalized interface, leveraging deeper hardware resources such as cameras and microphones to allow for richer integration into mobile use cases such as photo editing and real-time translation.
The industry's landscape has been reconstructed
These differences between web-based and app-based platforms have a significant impact on the global mobile AI application market. AI Chat Assistant continues to dominate at nearly 70% of MAU stocks, but leads and user scale are weaker than the web. Meanwhile, AI's search engine share has declined compared to web use, but AI image editing applications have skyrocketed to almost 10% shares as they fit more strongly into mobile photography and imaging scenarios and emphasize clear mobile users preferences.
On the consumer side, AI applications have shown strong penetration in assistants and office tools, but AI-driven creative and entertainment apps are struggling with stickiness and slowing growth. Assistant-type applications are already at the heart of the market, with deep voice interaction and smart assistant products growing rapidly and becoming the industry benchmark. Office and creative applications such as WPS AI and AIPPT.CN are also closely integrated into productivity needs and show strong growth. In contrast, lifestyle and entertainment AI apps are capable of short-term hype, but often lose users when novelty fades.
For enterprise applications, the key to successful deployment is to tailor features to your needs, ensure quantifiable ROI, and protect the right data and computing resources. Only when these three factors are in sync can a company unlock efficiency and sustainable value. Industries such as finance and healthcare already show great potential, sharing high value unique data, adequate budgets and clear demand characteristics, enabling meaningful benefits in automation, forecasting and decision support.
Today, Chinese companies are achieving returns from AI Foundation model applications, primarily through improved operational efficiency. Evidence across the finance, manufacturing and retail sectors demonstrates the impact of task automation, process optimization, and accelerated R&D. As technology matures and scenarios expand, the value of large-scale models is expected to shift from short-term cost reductions to strategic growth and revenue growth. This evolves from a pilot project to vertically integrated innovations that are deeply embedded in the core of your business.
Nevertheless, ensuring accurate alignment of enterprise needs and models remains a critical challenge. Research shows that 87% of companies believe that the current model is still lacking when dealing with extremely complex logical tasks. 62% reported a lack of standardized criteria for model selection. 50% indicates a discrepancy between model functionality and business needs. 39% say limited tweaking and deployment tools affect efficiency. Overall, achieving large-scale sustainable adoption requires further advances in accuracy, standardization, industry coordination and toolchain support.
Leadleo Research Institute is an original content platform for research into banks and businesses, and an innovative digital research service provider with nearly 100 senior analysts. You can contact the platform at cs@leadleo.com.
This commentary is the opinion of the author and does not necessarily reflect the views of Taketo's works.
Click to subscribe to Bamboo Works Weekly Free Newsletter here
