Morgan Stanley just slashed Baidu’s price target from $130 to $80, a 38% haircut that reeks of a paradigm shift. The market is not betting against Baidu’s technology; it is betting against the narrative that AI investment will magically translate into growth. This is not a blip. This is a narrative re-rating.
I have spent the last decade dissecting the gap between technological capability and commercial reality. I remember the Ethereum PoS transition in 2020: everyone was obsessed with energy consumption, but I was tracking the validator psychology, the shift in economic governance. The same analytical lens applies here. Baidu possesses a formidable tech stack: self-designed Kunlun chips, the Ernie LLM, PaddlePaddle deep learning framework, and a solid cloud infrastructure. On paper, it is a top-tier AI contender. Yet, the market is now pricing it as a mature, low-growth value stock. Why? Because the narrative of AI as a panacea for declining search revenue has been debunked by the numbers. The revenue forecast was cut by 1-9%, but the non-GAAP operating profit was slashed by 6-31%. That differential is the smoking gun. It tells us that Baidu is spending heavily on AI, but the return on that capital is not visible. The core insight is not that Baidu’s AI is bad; it is that the market is no longer willing to pay a premium for a story without a demonstrable profit engine. The narrative has shifted from "Baidu is an AI company" to "Baidu is a search company with an expensive side project."
Constructing new myths from the ashes of Luna taught me that narrative failures are often the result of a mismatch between the story being told and the underlying economic reality. The Terra/Luna collapse was not a tech failure; it was a narrative failure of "trustless code" without social consensus. Baidu’s current situation mirrors this. The narrative that "AI investment equals future growth" is being tested against the harsh reality of capital expenditure. The market is now deconstructing Baidu’s institutional legitimacy, mapping the regulatory and competitive landscape, and finding it wanting. The core of the problem is Baidu’s search business, which is increasingly under siege. The user base is not disappearing, but the monetization efficiency is eroding. Users are either searching and leaving, or they are using AI-generated answers that bypass ad clicks entirely. This is a liquidity fragmentation problem, but not of the DeFi kind—it is the fragmentation of user attention. Baidu’s search network effect, once a powerful moat, is being weakened by vertical apps like Douyin and WeChat, which are capturing high-intent commercial queries. The company’s response—pouring more capital into AI—is a classic case of treating a symptom as a cure. The AI cloud business, while growing, carries a lower gross margin and is project-based, not recurring. The SaaS-like attributes that the market craves are absent. The non-GAAP profit miss is a direct consequence of this: Baidu is failing to convert its AI cloud revenue into high-quality, recurring income.
The contrarian angle here is that the market’s pessimism might be premature, but for the wrong reasons. The sell-side is now baking in a 10x PE for 2027, implying zero growth and maximum AI cost. This is a value trap narrative. The real blind spot is the potential for AI to reshape Baidu’s core product. If Baidu can transform its search box into an AI Agent that generates direct transaction revenue—not just ad clicks—the narrative could flip again. But that requires a fundamental shift in product strategy, not just more GPU spending. The market is not seeing this because the data points are not there yet. The narrative is currently dominated by the "cost of AI" rather than the "value of AI." I am tracking the on-chain data of AI agent adoption in China, but that is a separate thread. For now, Baidu remains a company caught between two narratives: the old one, which is dying, and the new one, which is not yet born. The market is right to be skeptical, but it is also in danger of ignoring the possibility that the AI narrative could be rehabilitated, just as the Terra ecosystem narrative is being rebuilt from the ashes.
The takeaway is not to buy or sell Baidu; it is to understand that the market is shifting from a growth-at-any-cost narrative to a proof-of-execution narrative. The next 12-18 months will determine whether Baidu can build a new myth or become a relic of the internet age. The real question is not whether Baidu has AI, but whether it can build a product that people will pay for, not just search on. And that, my friends, is a narrative that has yet to be written.