Baidu is a Chinese technology company that dominates the country’s search engine market while rapidly expanding into artificial intelligence and cloud computing. It generated RMB 133.1 billion in revenue last year, making it one of the largest internet businesses in China. While its core advertising business faces stiff competition from social media apps, Baidu is currently transforming itself into a leader in generative AI for the Chinese market.
The investment thesis on Baidu is that its early lead in generative AI and cloud infrastructure will allow it to capture the transition of China's internet from simple search to AI-driven services. Its real asset is the ERNIE foundation model, which already handles over 1.6 billion API calls per day and is being integrated into every part of its software.
We believe Baidu is one of the most credible ways to own the AI transition in China, though it requires patience as the core advertising business navigates a weak consumer environment. The stock essentially offers a stable, cash-producing search business with a massive, unpriced call option on the future of Chinese AI.
Baidu's stock price has dropped significantly over the last few years and currently sits well below where it traded half a decade ago. The company is trying to reinvent itself as an artificial intelligence leader to find new growth, but it is currently struggling with tough competition and being placed on a restricted list by the U.S. government.
What does it do?
Baidu is a mature technology business that earns most of its money through digital marketing and is rapidly growing its AI-powered cloud segment. When users search for information on the Baidu app, businesses pay to display ads or links, similar to how Google operates. The company also rents out its computing power and AI software to other businesses through its cloud division, where it charges for storage, processing, and access to its ERNIE AI models. Customers range from small local shops buying search ads to large corporations building their own AI applications on Baidu’s infrastructure.
Where does revenue come from?
Baidu earns about 79% of its revenue from its core business, while the remainder comes from its majority stake in the streaming service iQIYI. The core revenue is split between online marketing, which accounts for the vast majority, and non-marketing services like AI Cloud and autonomous driving. Geographically, almost all of its revenue is generated within mainland China.
Revenue Breakdown
Revenue by Geography
Who are its customers?
Baidu serves over 1 billion monthly active users on its search platform and hundreds of thousands of enterprise customers through its AI Cloud. In the most recent year, its core business served a massive base of advertisers while its AI-enabled Baidu Wenku tool reached 94 million monthly active users. The company also operates Apollo Go, an autonomous ride-hailing service that provided approximately 826,000 rides in the first quarter of 2024. Its developer community for the PaddlePaddle AI platform has grown to 13 million people, showing deep roots in the Chinese technical ecosystem.
What gives it staying power?
Baidu has staying power because it controls over 70% of the search market in China, creating a massive data advantage for its AI. This dominant position makes it the default starting point for information in China, making its search data incredibly difficult for rivals to replicate.
Where is it headed?
Baidu is betting its entire future on becoming the foundational platform for artificial intelligence in China. Management is shifting away from being just a search engine to becoming a "Mobile Ecosystem" where users interact with the ERNIE bot for everything from shopping to work tasks. If this works, Baidu will move from selling simple ad clicks to providing the essential software layer for the Chinese economy.
Revenue growth has slowed as the core advertising business matures, but the high-growth AI Cloud segment is starting to move the needle. Total revenue for the last full year was RMB 133.1 billion, and while advertising was down slightly, cloud revenue accelerated to 26% growth in the most recent quarter. This shift shows a business successfully trading old, stagnant revenue for a new, faster-growing category.
Baidu is a consistent cash generator, though free cash flow can be volatile due to heavy spending on AI servers and data centers. It generated RMB 13.1 billion in free cash flow last year, which gives management plenty of room to fund research and development while still buying back shares. The gap between earnings and cash is mostly due to the high cost of building the massive computing power needed to run its ERNIE AI models.
The balance sheet is exceptionally strong, with a large net cash position that provides a massive safety net. Baidu carries roughly $37.5 billion in market value but sits on a mountain of cash and short-term investments that far exceed its debt. This financial strength allows the company to survive prolonged periods of weak economic growth in China without needing to raise new capital.
Baidu is a financially disciplined business that is effectively using its search cash to fund a new AI growth engine.
The AI Cloud business is the clear standout, with revenue growth accelerating to 26% in the most recent reporting period. This growth proves that Baidu is the preferred partner for Chinese enterprises looking to integrate generative AI into their operations.
Online marketing revenue decreased by 3% recently, signaling that competitors like ByteDance and Tencent are taking a larger share of ad budgets. If search advertising continues to shrink faster than the cloud business can grow, total profits will come under heavy pressure.
The Chinese internet advertising and cloud market is worth approximately $150 billion today and is expected to grow roughly 6% annually as it reaches maturity. While the market is massive, pricing power is under pressure as users shift their time away from traditional search and toward short-video platforms like Douyin. Baidu remains the leader in the search niche, but it is now forced to compete in the much faster-growing AI Cloud market where enterprise relationships are more important than consumer brand loyalty.
The Chinese tech landscape is brutally competitive, defined by "walled gardens" where giants like Tencent and ByteDance block Baidu from indexing their social and video content. This forces Baidu to rely on its own app ecosystem to keep users, leading to high costs for acquiring traffic.
ByteDance is the most dangerous threat because its Douyin app has become the primary search engine for younger users seeking product reviews or entertainment. This structural shift in user behavior is slowly eroding the utility of Baidu’s traditional search bar. Alibaba also presents a serious challenge in the cloud market, where it uses its massive e-commerce scale to underprice Baidu on basic computing tasks.
Baidu is losing overall attention share to video apps but is successfully holding its ground as the technical leader in high-end AI services.
Baidu's primary protection is its massive library of proprietary search data and the technical complexity of its ERNIE AI models. Its 70% share of Chinese search provides a unique feedback loop: the more people search, the better its AI models become, creating a high bar for any new competitor to clear. This data advantage is currently being repurposed to make its AI Cloud services more accurate than generic competitors.
The combination of 42% gross margins and steady cash flow proves that Baidu still has significant pricing power in its core niche. However, the lack of a dominant social network or payment system means it lacks the "lock-in" that Tencent or Alibaba enjoy. The moat is currently narrow because search is no longer the undisputed gateway to the internet in China.
The moat is stable as Baidu successfully trades its declining search dominance for a new lead in AI infrastructure.
AI Cloud grew 26% but online marketing revenue fell 3%.
Repurchased $669 million in shares during the most recent fiscal year.
Founder Robin Li maintains significant voting control and a multi-billion dollar stake.
Capital Allocation Track Record
Robin Li has shown remarkable strategic foresight by pivoting the company to AI early, but his team has struggled to defend the core search business against the rise of short-form video. While Baidu’s technical execution on the ERNIE models is world-class, the company has often been late to respond to shifts in consumer behavior, allowing ByteDance to capture the next generation of users. The management's ability to turn technical wins into stock market returns has been mixed, often overshadowed by broader economic concerns in China.
The primary governance risk is the high level of control held by Robin Li, which makes the company's future entirely dependent on his personal vision. While this allows for long-term bets that quarterly investors might hate, it also means there are few checks and balances if a major strategic bet like autonomous driving takes longer than expected to pay off. There is a solid bench of technical talent, but Baidu remains very much a founder-led organization with the key-person risks that entails.
We expect revenue to grow from $132B in FY2026 to $186B in FY2031 (~7% CAGR), with EPS growing from $52.83 to $92.81 (~12% CAGR). Growth is driven by the steady expansion of AI-powered cloud services and the recovery of digital marketing spend in the Chinese market. Profitability improves as the company moves past heavy initial investments in autonomous driving and AI infrastructure. EPS grows faster than revenue because profit margins are widening and the company is buying back its own shares. Operating margin expected to reach ~18% by FY2031.
AI Cloud becomes the primary infrastructure for Chinese enterprise software. As Chinese companies race to integrate AI, Baidu's lead with the ERNIE model could turn its cloud division into a high-margin monopoly for AI services.
Apollo Go reaches profitability through fully driverless operations at scale. If Baidu can remove the safety drivers across its entire fleet, the robotaxi business transforms from a massive cost center into a high-margin transportation utility.
ERNIE integration revitalizes search through higher-quality AI-generated answers. By providing direct, high-quality AI answers, Baidu can increase user engagement and charge advertisers more for more valuable, intent-driven leads.
Geopolitical tensions restrict access to the high-end chips needed for AI. If Baidu cannot secure the advanced processors required to train its next-generation models, its technical lead over rivals could evaporate.
ByteDance and Tencent build superior AI models integrated into social apps. If AI search becomes a feature inside WeChat or Douyin, Baidu's standalone search utility could become irrelevant to the average consumer.
Prolonged economic weakness in China permanently depresses digital advertising budgets. A structural slowdown in Chinese consumer spending would starve Baidu of the search cash it needs to fund its AI and robotaxi bets.
Below is our estimate of current and future fair value, with detailed reasoning and assumptions. Fair value is a judgment, not a fact, and other analysts will likely land on different numbers. Use it as one data point in your research, and apply your own discretion in any investing decision.
We use a Forward P/E approach based on next-fiscal-year earnings (FY+1). This framework is most appropriate for Baidu because it has become a "value-tech" play where massive GAAP (Generally Accepted Accounting Principles) earnings are the primary signal, but those earnings are heavily discounted by the market due to geographic and regulatory risks. P/E allows us to explicitly model this "China Discount" by adjusting the multiple relative to US-based peers.
Multiplying the FY2027 EPS of $58.98 by a 2.8x multiple gives a per-share fair value of $165. Our 2.8x multiple sits significantly below domestic peers like Alibaba (9.2x) and JD.com (8.1x) to account for Baidu's higher exposure to the competitive search-ad market and its earlier-stage AI pivot. The EPS basis of $58.98 is pulled directly from the deterministic projection for FY2027, representing a steady but realistic growth path for the core business and cloud initiatives.
A peer-anchored Price-to-Sales (P/S) cross-check produces a fair value of $167, almost exactly matching our $165 P/E-based answer. We applied a 0.4x P/S multiple—consistent with the low end of Chinese large-cap tech—to the projected FY2027 revenue of $141.8B. This results in an implied market capitalization of roughly $56.7B, which, when divided by 340 million shares, confirms the significant undervaluation relative to the current $110 price. Both methods highlight that the market is currently pricing Baidu as if it were a distressed utility rather than a growing AI leader.
We're assuming the AI Cloud segment continues to grow at a 35% compound rate through 2028. While it grew 79% in the most recent quarter, we expect natural deceleration as the business matures and domestic competition for GPU (Graphics Processing Unit) cloud capacity scales. This growth is central to offsetting the slow decline of the legacy search business.
We're assuming the "China Discount" remains a permanent fixture of the valuation. Historical tech multiples of 15-20x are no longer applicable to Chinese internet stocks; our model assumes a ceiling of 4x even in a bull case. This reflects the structural risks of the ADR structure and the unpredictable regulatory environment in the People's Republic of China.
We're assuming Baidu maintains a net cash position to fund its autonomous driving expansion. The company currently holds roughly $17B in total liquidity (cash plus short-term investments), which provides a critical safety buffer as the Apollo Go robotaxi service expands into 27 cities. This liquidity effectively floors the stock's value even during periods of earnings volatility.
The single biggest risk is a "valuation trap" where Chinese ADRs continue to trade at distressed levels regardless of fundamental growth. Even if Baidu hits its $92.81 EPS target by 2031, continued geopolitical friction or regulatory crackdowns could keep the multiple suppressed at 2x, preventing the fair value from ever materializing. Watch for any move in the "China Equity Risk Premium" that pushes the broader Hang Seng Tech index toward new 5-year lows.
Bear case ($88): AI Cloud revenue growth slows below 30% as domestic competition from ByteDance and Alibaba intensifies; or Geopolitical tensions trigger a broad sell-off in Chinese ADRs (American Depositary Receipts), compressing the multiple to 1.5x.
Bull case ($248): AI Cloud and Apollo Go (autonomous driving) achieve segment-level profitability, prompting a re-rating to a 4.2x multiple; or Baidu successfully transitions over 60% of legacy search users to its "agentic AI" platform, stabilizing advertising margins.
Clearthesis wrote this report from 34 sources, including SEC filings, industry research, and recent news.
How did you like this thesis?
Your feedback helps us make reports better for you
© 2026 Clearthesis.ai · Report generated on June 23, 2026
This is an AI-generated analysis for informational purposes only and does not constitute financial advice. Data and analysis may not reflect recent developments if viewed significantly after the generation date. Always conduct your own due diligence before making any investment decisions.
The market is leaning bullish because Baidu is successfully pivoting from traditional search to becoming the foundational AI provider for China. Its ERNIE model has gained significant traction, and recent regulatory approvals for autonomous robotaxi services in Switzerland demonstrate that its AI technology is now capable of operating on global standards.
Skeptics think that Baidu remains a dangerous bet due to its entanglement with the Chinese state and geopolitical friction. The Pentagon's decision to label Baidu as a military-linked firm introduces a constant risk of sanctions or exclusion from international markets that could paralyze its long-term growth ambitions.