Cheetah Mobile Inc.

Good day and welcome to the Cheetah Mobile first quarter 2026 earnings call. All participants will be in listen-only mode. Should you need assistance, please signal a conference specialist by pressing the star key followed by zero. After today's presentation, there will be an opportunity to ask questions. To ask a question, you may press star then one on your telephone keypad. To withdraw your question, please press star then 2. Please note this event is being recorded. I would now like to turn the conference over to Chita Mogul, Investor Relations, Helen. Please go ahead.

Thank you, Alfredo. Welcome to Chita Mogul's first quarter 2026 earnings conference call. Lisa today, our company's chairman and CEO, Mr. Fu Sheng, and our company's director and CFO, Mr. Thomas Jennings. Following management's prepared remarks, we will conduct the Q&A section. Please note that the management's praise will be presented by an AI agent. Before we begin, I refer you to the phase helper statement in our early training, which also applies to our conference call today. as we will make forward with these things. At this time, I would not like to turn the conference over to our Chairman and CEO, Mr. Fusheng.

Please go ahead, Fusheng. 2026 remains an important transition year for Chiba Mobile. We are continuing to evolve from a traditional internet company into a company focused on AI-enabled applications for AI agents and robotics. More importantly, We believe we are gradually moving from capability building into early-stage commercial validation. Our focus is not only on developing the high capabilities, but on turning these capabilities into practical products for real business scenarios, helping customers deliver better ROI. Starting from this quarter, we are separating our advice and other into an independent reportable segment. In the first quarter, revenue from robotics and others business increased 176% year-over-year to 51 million RMB, approaching 20% of total revenue. And at the same time, adjusted operating loss from this segment narrowed by 57% year-over-year. Customer demand remains strong, and we expect robotics and others revenue to grow strongly in 2026. In Q2, our robotics and other revenue will continue growing on both year-over-year and quarter-over-quarter basis. Today, our robotics business mainly focuses on commercial scenarios with real customer demand and clear launch and value, including reception, guided tours, and intelligent service applications. Our smart personal mobility is another important step for us. This product extends our robotics and AI capabilities into personal mobility and healthcare-related scenarios. More importantly, it further validates that our robotic platform can expand beyond commercial service robots into broader consumer applications. We are encouraged to see recognition from leading industry partners. During the second quarter, we started initial product shipments to a top mobile designer and manufacturer of mobility products as well as the leading alien mobility scooter manufacturer in China. We are seeing encouraging early market feedback and initial commercial traction. Moving to our agents, we're seeing strong customer adoption. We worked closely with Google Cloud and AWS, helping enterprises serve international markets, access AI models, and use multi-cloud environments more efficiently in one Q2. ai infrastructure services as a part of global enterprise service revenue increased 68 percent year-over-year contributing 18 percent of total revenue daily energy token usage has increased more than 20 times since january of 2026 exceeding 400 billion in may we expect this revenue growth to continue we also kept building easy goal So early, but we believe it will help customers deploy AI apps and boost productivity. The two-plus growing businesses, namely robotics and others, as well as cloud and AI infrastructure, already accounted for 38% of our first quarter revenue, and we expect their revenue growth and revenue contributions to continue to grow. from our advertising agency business within the global enterprise services segment was affected by policy changes from certain overseas advertising platforms. We believe this revenue decline was primarily driven by external factors rather than changes in customer demand. This was the primary reason the first quarter. Our internet services business continues to provide important profit and cash flow support for the company. In the first quarter of 2026, our internet service business generated approximately 15 million RMB in adjusted operating profits.

Excuse me, there has been an interruption. Just one moment, please.

profit and cash while as agency revenue was hit by policy changes which impacts our financial results in the near term it is a stronger base for growth moreover our us dollar 186 million that also supports our ai agents and robotics thank you thank you fu shang hello everyone and thank you for joining us unless otherwise it is for financial figures

presenters in RMB. During the first quarter of 2026, we continued focusing on operating discipline, improving revenue quality, and maintaining financial flexibility as we invest in AI and robotics initiatives. Postal revenue remained relatively stable year over year at 259 million RMB during the quarter, while internet service revenue declined due to convenient weakness in online advertising All of our revenues continue to improve. Within the internet service segment, revenue from internet value added services continue to grow steadily at 8.2% year-over-year, contributing 72.8% of segment revenue. Within a larger portion from internet value added services, our internet service revenue is becoming increasingly predictable. More importantly, the internet service business remains profitable and continues generating stable cash. which provides an important financial foundation for our long-term AI and robotics investments, turning to our robotics and others segment. Starting from this corner, we begin reporting the robotics and others business as a separate segment to present the operating progress of this business. Historical results previously reported on the AI and others are now presented as robotics and others, as well as global enterprise services. During the first quarter, revenue from robotics and others increased significantly year over year, with revenue increasing 175.9% year over year to 51.2 million RMD, accounting for 19.8% of total revenue. Suggested operating loss from this segment narrowed by 57.1% year over year, reflecting continued improvement in operating efficiency and commercial execution. Starting through global enterprise services, this business remains strategically important to the company. In addition to profitability attribution, it provides us with valuable enterprise customer relationships, overseas operating experience, and real-world deployment scenarios for AI-related services. During the quarter, Revenue from the advertising agency business was affected by policy changes from overseas advertising platforms, which impacted year-over-year segment revenue performance. However, revenue from our cloud and AI infrastructure services business increased by 68.3%, supported by increasing advertising demand for AI-related cloud and token management services. Moving to profitability. Operating loss was on the 28.3 million during the quarter, compared with 26.5 million RMB in the same period last year. Increase mainly reflected lower profitability on the internet and global enterprise services business, following revenue declines in online advertising and advertising agency services, as well as our continued investment in AI and robotics initiatives. More importantly, the internet service and global enterprise services business remain profitable during the quarter. The internet service business generated approximately 15.2 million RMB in adjusted operating profits, while global enterprise services generated approximately 13.8 million RMB in adjusted operating profits. We also maintained a strong balance sheet. As of March 31st, we had approximately 186 million United States dollars in cash and cash equivalent, as well as over 100 million United States dollars in long-term investments. We believe our financial position provides sufficient flexibility to continue investing in AI and robotics with a disciplined and sustainable approach. Looking ahead, our financial priorities remain consistent. Thank you. maintaining operating discipline. B, improving revenue quality and operating efficiency. C, supporting long-term investments by preserving financial flexibility. Overall, we believe the company continues moving toward a more sustainable and balanced operating structure as our AI and robotics businesses gradually scale. Thank you. We are now ready to take your questions.

Operator, now we are able to take questions.

We will now begin the question and answer session. To ask a question, you may press star then 1 on your telephone keypad. If you're using a speakerphone, please pick up your handset before pressing the keys. If at any time your question has been addressed and you would like to withdraw your question, please press star then 2. At this time, we will pause momentarily to assemble our roster. The first question comes from Thomas Chong with Jefferies. Please go ahead.

晚上好,谢谢管理层接受我的提问。 最近我们看到市场上越来越多 关注机器人机能化能力的提升。 很多业内人士认为 真的有价值的不仅是训练阶段的数据, 更是在分学部署过程中, 持续产生回流并不断优化系统的数据。 想问热道在过去几年, 已经是在多过商业场景中长期运营机器人。 从你的观察来看, 这些运行过程中不断积累的动态的数据, 对于机器人能力的提升的重要性, 提升在哪些方面呢?

我们对机器人行业都有非常高的憧憬, 认为这个 Today, AI's agent ability has been enhanced, and robots should be able to achieve various behavior abilities very quickly. But in fact, I don't think so. Because the development of AI agents, including the development of large-scale models, is actually based on the development of the Internet for two or three decades. In other words, the Internet itself has formed the basic training data for large-scale models. It is a very high-quality data machine. The biggest problem in the robot industry today is that there is no data. Today, many manufacturers are trying to use training data, including data transfer, simulation training, and so on. But there is a very serious problem. The physics world is much more complicated than the laboratory environment and the simulation environment. So today, whether it's data transfer or . . . I also agree with what you said. The data that is continuously generated in the real deployment environment is very important for the robot industry. From our own experience, for example, I will give two examples. One is that we have the ability to communicate with each other in different environments. In fact, we have done some data optimization and It's related to training. So the effect of our delivery of the robot, including the reception, and the performance in the industry today are all... I can't brag, but it's really in the front row. It has our own reputation in the industry. Another example is the mobility of the robot. A very simple indoor navigation from point A to point B of the robot is similar to a small, low-speed unmanned vehicle. How to use... cheap chips and sensors to achieve very good passability and barrier capability in different environments. In fact, this has to be done on the basis of a large number of such data. We recently launched a smart wheelchair. We just talked about it. In fact, we officially started mass production in May and started to push it into the market. Now it looks like it's pretty good overseas, especially in Europe. In fact, this kind of ability to be able to do such a product on a traditional wheelchair can assist the driver, similar to the ability to assist the driver. Many manufacturers, including entrepreneurs, want to do it. But if you want to make a prototype and really do it in many environments, you can do it very well. In fact, it takes a lot of effort. This is what we have deployed in many environments for many years. 在无论是什么样的地面条件 比如地毯 地板 对吧 不同的包括要增加墙壁的反射 这些东西都是经过大量积累的 还有在不断的算法优化 根据实际的场景做算法优化 所以我们这款轮椅 真正能够实现较低成本的 高度的可辅助的这种驾驶性能 也在市场上得到了很好的反馈吧 So we think that the ability to develop today's so-called universal machines, I always think that it has to be accumulated step by step. A better way is to get data from this actual scene, and then a little bit of accumulation to gradually realize the ability one by one. It is possible to achieve a better commercialization. So we highly value this point. In fact, some people have also asked why we want to push products like smart wheels. Because in our eyes, it is essentially Operator, can we move to the next question?

Yes, the next question comes from Vicky Wei from City Securities. Please go ahead.

Vicky, I run the line.

Okay, we'll go to the next question. The next question comes from Lydia Lin from Morgan Stanley. Please go ahead.

Thank you for accepting my question. What I want to ask is because in the past few years, the market's attention to the AI industry has mainly focused on this model capability and model manufacturers. Thank you.

This is a very big industrial observation. My opinion is just a reference. I think in the long term, the ultimate value of the industrial chain will still come from the application layer. Although in the past two years, especially this year, the large model and the infrastructure layer, everyone is particularly concerned. Because I think fundamentally, There is a saying that in the carbon economy, the bottom-up construction is not enough. This construction is very scarce in a period of time. So there will be such a big gap. So at this stage, the value chain is definitely in this regard. But I want to say the first one, why do I think it is not the model layer? Because although the model is very fierce in all kinds of competitions, but now we see the situation is that the gap between the model Taiwan Taiwan . . . Even if it is not a top model, when it is matched with some daily tasks, it has actually achieved a very good performance. For example, some open source models in China this year have been significantly improved in volume. I think the core is that its cost performance is very good. The completion of some tasks is already very high. So even I think that in the future, all kinds of professionalized models will continue to emerge. Of course, it will take some time. We don't know about the second basic infrastructure layer. China China China China China AI can actually repeat almost all applications. So today's application layer has a huge opportunity. Whether we are doing a robot-like industry, we have been doing it for a long time. But we are still very determined to look at it. When the ability of the model continues to improve, when the use of the robot is more widespread, there are many rumors that it may be a bigger industry than the car. And there are also a lot of opportunities at the software level. Let's not talk about it here. Even if we look at some big model companies today, their valuation is very high or very outstanding, right? In fact, it has also truly taken root in a certain application. For example, as I said, its programming, right? Its cloud code is rising. In fact, it is also an application. Its application is the application of code writing. It does the code writing application well enough, and not just provides API to let you call it, but its agent does it well enough. Including this year's rising like Open Cloud, right? We also made products like Easy Cloud. 所以我觉得应用层还是有非常广阔的填空和机会的。 好,谢谢。 Operator, can we move to the next question?

The next question comes from Vicky Wei with Citi. Please go ahead.

管理层,晚上好,谢谢接受我的提问。 我也想请教您关于机器人行业的, 就是市场对于未来的竞争国际有很多讨论, 我们对机器人行业理解来说 我认为短期内或者可见的三五年内出现一款

The possibility of particularly common robots is very, very low. This is not only due to the so-called model capability, but also due to the entire hardware industry chain. Like the update of hardware industry chain, in fact, its speed is relatively slow. And it only has some of the most basic materials and materials. The logic of materials and materials is at the bottom. So today I will think that the core of the future robot industry is closed. 我们自己独有的无论是经验还是数据。 And then I just answered the first question, that is, we can optimize it. And this optimization also makes this product able to truly meet the needs of users with better cost-effectiveness. The robot industry is very popular, but when it really falls in the business, in fact, customers do not care whether you are a robot or a machine or a person. What they care more is the cost-effectiveness, ROI, input and output. This has been clearly reflected in our operations in the past few years. So even if you see a lot of experienced robots in the media, you will find that they are very, very few in the actual scene. And without going through the actual scene, I would like to reiterate that the complexity of the operation of robots in the physical environment, whether it is action or work, is actually much higher than the automatic driving of cars. So in this, In this very complex situation, I think it is more important to be able to form a vertical, penetrating point in the actual use of the scene, in the operation of the scene, and in the customer network. It is much more important than the so-called universal machine and model of a defect. Because today, I don't think the defective model and machine can be 好,谢谢。 Operator, please move to the next question.

Yes, the next question comes from Nancy Liu with JP Morgan. Please go ahead.

喂,谢谢管理层接受我的提问。 我们看到最近基础模型能力吃铜与API成本持续下降, Thank you.

我觉得这是一个很犯的问题啊 我觉得企业阶级产品最终互中和 应该来自于对这个用户需求的深度的理解 和对这个行业的深度的这个理解 然后形成了一个超高化的组织能力 因为今天你讲这几点 我觉得也都是现实 就是模型本身的这个能力 像看上去此消彼长 And then the cost-efficiency ratio is being raised more and more. So what is the essence of today? It is that it actually allows companies to save a lot of money that has been spent on non-business survey, user survey experience, right? To save it to the real survey of user needs. So the real deposit box comes from the subtle survey of your user needs, and then quickly launch your product and service. and improve your products and services. So we often talk about this AINAT5 organization. Its essence is to use AI to rebuild the organizational process within the enterprise, and to be able to quickly and efficiently implement the operation of the enterprise, to be able to more efficiently and quickly launch our own products and services. For example, if you pay attention to the observation, we have launched a variety of product services in the past year. It's also much faster, but our investment in the cost of this development is from the cost point of view, we have reduced a lot Although there is room for improvement, this is just an example, then you are so fast to promote products and services, right? Where is your real container? Is it from the needs of users? That is, you can really find the needs of users and quickly launch and quickly follow the needs of users . . . . . . . . . Thank you.

Thank you. Operator, please move to the next question.

The next question comes from Gigi Zhao with GF Securities. Please go ahead.

Hello. Hello, Mr. Guan. As you mentioned, there is a lot of investment around the AI project. We would like to understand that a large number of AI projects are still dependent on customized development and artificial services. There is still a certain distance between the standardized delivery mode of traditional SaaS products and the traditional SaaS products. As the AI agent's capabilities keep improving, do you think AI applications will eventually move towards a more standardized SaaS model, or will they maintain the hybrid model of software and service for a long time?

What are the key changes in this process? I think the reason why there are so many customized development and AI services today is because AI is still in its early stages. Although we all see in the circle of friends or in various media that AI is a full-fledged AI, in fact, the vast majority of people still don't have enough understanding and use of AI. I think only a few people can really use AI well today. So this is a break. That is to say, today's AI project is in a company with history. then it has to do custom development and artificial service for this history. And the traditional SaaS has been developed for many years. It has concentrated a lot of things in the code. So it seems to be standardized in many cases. I think as we gradually understand AI, the whole, I should say the whole, the entire industry is constantly familiar with the application of AI. The proportion of service mode will continue to decline. In our company, the whole team is using AI to write code, and some of our internal systems are also using AI. Business departments write directly, instead of relying on SaaS software and service departments. So I think the most critical change in this process is that, on the one hand, I think the model capacity is constantly increasing. Today, for example, we have a very important feeling this year, which is that today the business department is writing some internal software and services. uh uh traditional standardized SaaS model, my point of view is that it will be broken. So what we provide our customers today is no longer this kind of traditional delivery type of service, but more through training, training for our customer employees, as well as the assessment of AI capabilities to help them do the transformation of AI organization. 我觉得这个变化是最关键的,也就是说企业要根据AI去改变自己的组织结构和对员工的要求。 谢谢。 好的,感谢副总的精彩的解答。 我这边是国联民生证券的赵志远,我这边问题了解。 好,谢谢。 Thank you, Joglin.

Operator, please move to the next question.

Thank you. The next question comes from Hupeng Tao with Kutai Haitong. Please go ahead.

Thank you.

I think I have already expressed my views on humanoid robots very clearly in the media. I believe that humanoid robots are hard to achieve in the next three or five years, not only in terms of performance, but also in terms of commercialization. That is to say, whether it is in the factory, in the service industry, or even in the family, I think humanoid robots should be said to be very, very difficult. Yes, we ourselves have a wheelchair robot and a mechanical arm. I think my point of view has always been like this. I think the robot should be specialized from one to one. It's called a vertical machine. It keeps growing and collecting data. When it develops well enough, it will continue to integrate and gradually move towards a more general machine type. As for the dual group itself, I don't think it's necessary in most scenarios. There is no need to increase such cost and complexity, including its reliability. So this is my point of view. And we have also reiterated that what we are most concerned about as a robot is commercialization. Is it really able to be paid by the market? And is it really the main market payment? And not just some project landing or some high-end projects. So I think it should be a wheelchair robot. In the future, it will gradually be paired with B. Operator, please move to the next question.

Thank you. The next question comes from Guang Zhang with Guohai Securities. Please go ahead.

Okay. I would like to ask, family service robots are considered to be the largest market in the robot industry for a long time. But at the same time, it is also the most complex demand and the most challenging scene. Leopold has also launched its own smart wheel product this quarter. In the next two to three years, what is the most likely way for home robots to make a breakthrough? And what are the real big-scale challenges in the industry that need to be overcome? Thank you.

Yes, in fact, home robots are a very bad concept. If you want to talk about home robots, the one that has made a breakthrough is the sweeper robot. It's also called a robot, right? But if we consider it as being able to do a more繁重任务的, I think the first one is that we did the smart wheelchair because in our point of view, the smart wheelchair is a robot. It's just that before, our robots delivered goods, right? The smart wheelchair, in fact, you can understand it as delivering people. I think the first one in the family is the mobility, from A to B's mobility. Then the second one is to increase some additional capabilities on this mobile capability. For example, the sweeper machine is to increase the sweeper capability, right? What we see now is companionship. And it's to help you achieve some family control. Control it in a voice way, right? Merge it into the robot. Help you make some plans. Do a good companion. It's all one direction. 我认为两到三年内是不可能实现的。 这个因为我们自己有机械币公司, 我们的机械币在很多场景, 无论是工业场景, Today, even if you grab a cup of success rate, Right? 没有一家能做到百分之百 在一个 哪怕是在一个厨房环境当中 去抓取一个杯子 但是作为家庭应用来说 你抓取这个杯子 只要是 哪怕是99%的成功率 你也会每段时间摔杯子 对吧 它带来的这个负面效果是非常大的 更不要说如果是人行进家庭 它有摔倒的问题 砸到东西的问题 砸到人的问题 还有就是它的质量的可靠性问题 对吧 一个家电你买回去大概 The quality of this few years is no problem, right? Today, if you really want to be a complicated robot, in a long period of time, the quality can guarantee that it is not bad. Today, it is a problem for many robot companies. So I think family robots, we should be a little more realistic, right? Our point of view is that this can really achieve family companionship, can achieve some old people, disabled people's movement. I think this is a very good way to break through. Thank you.

Okay, operator, please chat if you have further questions. Thank you.

Thank you. Seeing there are no further questions, this concludes both our question and answer session and today's conference. Thank you for attending today's presentation. You may now disconnect.