Another case of AI devouring software? Google's Nano Banana goes viral, MEITU's stock price plummets

The explosive popularity of Google's new AI model, Nano Banana, has sparked deep concerns in the market about whether "AI will devour application software." However, Morgan Stanley analysts believe that this concern is overstated.

On September 2, Morgan Stanley's research report pointed out that Meitu's growth trajectory has not been affected by the AI model Nano Banana, and its real value lies in providing "last-mile" solutions that basic AI models cannot reach.

The report emphasized that Meitu has built a solid moat through its deep accumulation in vertical fields, proprietary data, and focus on core paid features. In addition, the company maintains an open and neutral stance towards all AI models and has seamlessly integrated new technologies like Nano Banana into its products.

Morgan Stanley reiterated its "Overweight" rating on Meitu, with a target price of HKD 15.70, believing that its long-term value remains solid. It is expected that revenue will reach RMB 4.021 billion in 2025, grow to RMB 5.120 billion in 2026, and further increase to RMB 6.771 billion in 2027.

Market Panic Triggered by Nano Banana

Recently, the new image AI model "Nano Banana" released by Google quickly became popular on Chinese social media.

This model was launched on August 26 and, with its powerful ability to transform photos into 3D model images through prompts, it ignited public opinion on August 30.

This event immediately triggered investor concerns—whether increasingly powerful basic AI models would "erode" the core business of applications like Meitu.

Affected by this panic, Meitu's stock price plummeted by 14% on the trading day following the news, while the Hang Seng Index rose by 2% during the same period.

It is worth noting that this is not an isolated case.

Since the beginning of this year, whenever heavyweight AI models like GPT-4o or Google's Veo3 have released updates, Meitu's stock price has experienced similar fluctuations.

This reflects the general anxiety in the market regarding the survival model of application-layer software in the AI era.

"Last Mile" Value is Hard to Replace

Morgan Stanley analysts emphasized that in the AI era, the value of application software lies in providing "last-mile" services to optimize outcomes, which is something general AI models cannot fully achieve.

The "last mile" in the visual industry is particularly complex for two reasons:

• Fragmentation of scenarios and diversity of needs: User needs vary widely, from beautifying personal photos, generating advertising images, to poster design, each scenario has unique standards and workflows

• Subjectivity of personal preferences: Aesthetics and visual preferences are highly personalized, making it difficult to accurately capture them through standardized prompts, and cannot be perfectly realized by general AI models.

It is precisely these two characteristics that determine the difficulty for general AI models to provide an end-to-end perfect product that meets all user needs.

Therefore, application software that focuses on specific scenarios and optimizes user experience has not only maintained its value but has become increasingly prominent.

Advantages of MEITU: Scenarios, Data, and Core Payment Points

Morgan Stanley believes that MEITU has established significant advantages in maximizing the value of the "last mile" due to its long-term accumulation.

First, the company's business focus has accumulated deep scenario awareness. MEITU focuses on non-professional leisure and non-professional productivity scenarios, further refining the latter into areas such as e-commerce design.

This focus allows it to deeply understand user pain points in specific scenarios and develop highly optimized workflows to enhance user experience.

Second, massive vertical data builds barriers. Through partnerships with Alibaba and long-term operation of its own products, MEITU has accumulated a vast amount of high-quality vertical data, especially in portrait beautification and e-commerce design.

MEITU utilizes this proprietary data to conduct "post-training" on top of third-party foundational models, creating smaller models that outperform general models, with results that are closer to user needs.

Finally, a clear business model: Users pay for "core functions." MEITU's features can be divided into "fun features" (such as AI emoji generation) and "core functions" (such as AI facial retouching).

The former may attract users to try the product, but the latter is the fundamental reason driving users to pay for subscriptions. General AI models can imitate the former (such as the 3D model of Nano Banana), but it is difficult to penetrate the high-quality "core function" area that MEITU provides.

In addition, MEITU adopts a neutral strategy for AI models, using AI model container technology to quickly integrate the capabilities of different AI models into its products.

Before the widespread dissemination of 3D model images, MEITU had already integrated Google's Nano Banana into its overseas products, including MEITU XiuXiu, BeautyPlus, and RoboNeo