NASDAQ: SERV

Our core technology originated in 2017 as a specialized project within Postmates Inc. (“Postmates”), an on-demand delivery company in the United States. Following the acquisition of Postmates by Uber Technologies, Inc. (“Uber”) in 2020, Uber’s leadership approved the contribution of the intellectual property and related assets associated with this project to our company in February 2021.

As of December 31, 2025, Serve’s fleet consisted of over 2,000 sidewalk delivery robots. We maintain platform-level integrations with food delivery platforms, such as Uber Eats and DoorDash, that enable our robots to transmit real-time presence and status information and to receive delivery requests associated with customer orders placed on those platforms.

In January 2026, subsequent to the period covered by this Annual Report on Form 10-K, we completed the acquisition of Diligent Robotics, Inc. (“Diligent”), a provider of AI-powered autonomous robot assistants for the healthcare industry. This acquisition extends our autonomy platform into indoor environments, initially into hospitals, and broadens our addressable market into healthcare and other commercial sectors. We expect to leverage Diligent’s technology alongside our existing platform to enhance product capabilities, accelerate deployment opportunities, and support scalable, recurring revenue growth through Diligent’s established commercial relationships with healthcare systems and institutions.

Tailwinds for Automation

Despite technological innovations of the past few decades as well as increased adoption of online commerce and home delivery, last-mile delivery has remained costly. While an ever-growing share of consumers is shopping online and demanding faster deliveries, a number of factors have contributed to keeping last-mile costs high:

•labor shortages caused by the aging population have led to wage inflation;

•on-demand delivery companies in many jurisdictions are facing regulatory pressures to classify gig workers as employees, which would in turn increase labor costs; and

•cities across the United States have introduced maximum limits on how much delivery platforms can charge restaurants and merchants, highlighting the need to lower underlying delivery costs.

Labor cost inflation and regulatory pressures serve as tailwinds that are expected to accelerate the adoption of automated robotic last-mile delivery. Similarly, the healthcare industry is experiencing significant labor shortages, particularly among nursing and clinical support staff, which has created demand for autonomous solutions that can perform routine logistical tasks and free clinical staff to focus on patient care. While labor costs typically increase over time, hardware and technology costs typically decrease. We expect the cost of cameras, GPU processors, electric motors, batteries, and advanced sensors such as LiDAR to continue to decrease, while mobile networks are expected to become faster and more reliable with higher bandwidth and geographic coverage, which we believe will lead to a downward trend in the cost of building and operating robots.

Robot Operations

Our robots start each day at a central depot located near their operating area. The fleet is monitored through mobile connectivity and video streaming by remote human supervisors who can assist robots when necessary, such as at intersection crossings or when robots are unable to navigate certain conditions. On less frequent occasions, if a robot requires physical assistance, such as when a robot is too low on battery to return home or if it has been damaged, a nearby team member is dispatched to repair or return the robot.

Throughout the day, each robot receives a series of delivery orders from partnered merchants and delivery platforms. Upon acceptance of an order, the robot navigates to the pick-up location, waits outside, and notifies merchant staff, often through their existing delivery tablets or point-of-sale devices. Once the merchant staff load the package into the robot, it navigates to its drop-off destination. Upon arrival, customers meet the robot at the curb, unlock its cargo using their delivery app or on-screen instructions, and retrieve their package. At night, robots return to their central depot to be recharged, maintained, upgraded when necessary, and prepared for next morning’s deployment.

Serve’s third-generation robots are designed to carry more goods, enable more deliveries, and further reduce the cost of delivery thanks to higher speeds and increased battery capacity. They can reach up to 11 miles per hour (18 kilometers per hour) and travel as far as 48 miles (77 kilometers) on a single charge. They also feature an expanded 15 gallon (61 liter) cargo bin that holds four large 16-inch pizzas. Their new drivetrain is equipped with suspension to drive smoother and faster while protecting food quality, and their improved water resistance expands the robots’ ability to maneuver confidently in a wider range of weather conditions.

Importantly, these hardware and software enhancements support Serve’s commitment to safety on the sidewalk. In addition to market-leading safety capabilities, including fail-safe mechanical braking and autonomous collision avoidance, the third-generation robots feature enhanced emergency braking.

Following our acquisition of Diligent in January 2026, we also operate Moxi robots in hospital environments to assist clinical staff with logistical tasks. Moxi robots operate throughout hospital facilities, navigating hallways, using elevators, accessing secured areas via badge readers, and entering patient care areas to deliver medications, lab samples, and lightweight equipment and supplies. Moxi robots operate autonomously within hospital environments, using onboard sensors and AI to navigate dynamic spaces.

All of our robots are equipped with a number of security features designed to protect their cargo, data, and intellectual property from unauthorized access. The robots’ secure cargo compartment can only be unlocked through the delivery app’s interface or using a security code entered on the touch screen. Additionally, through the use of encrypted communication protocols and data storage as well as secure authentication methods, the data and software onboard each robot is protected from unauthorized access. All of our robots that operate outdoors are also equipped with redundant real-time location tracking systems as well as alarm and communication features to deter attempts at vandalism and enable quick recovery of assets in the unlikely event of theft.

Impact of Robotic Automation

If automation decreases the cost of last-mile delivery and leads to increased adoption, we anticipate opportunities for impact over the long term, such as:

•Reduced Greenhouse Gas Emissions: We believe the use of robots and drones will reduce overall emissions caused by the use of large vehicles for moving small packages.

•Lower Delivery Costs: While all automated delivery robots still require a certain amount of human involvement (e.g., loading and unloading, maintenance, remote supervision), we believe labor is leveraged more efficiently, resulting in more deliveries and tasks per unit of human effort. Just as automation has done in the past, we expect a continued reduction of delivery costs over time.

•Increased On-Demand Adoption: On-demand services are largely luxuries afforded by affluent consumers today. We believe reducing delivery costs could make home delivery services affordable to more people.

•Easier Reverse Logistics: We believe reducing the cost of last-mile transportation will also encourage increased adoption of reverse logistics applications (e.g., more convenient package returns).

•More Local Commerce: We believe that increased adoption of home delivery will result in more commerce for local businesses.

•Increased Local Jobs: We believe increased delivery and local commerce activities resulting from a reduction in delivery costs could lead to more local jobs, ranging from increased staffing of local businesses, logistics operators who enable automated delivery networks to function, and human couriers who perform deliveries that automated services cannot perform.

•Higher Delivery Quality: Like most automated systems, robots are less prone to certain types of errors. For example, customers of food delivery platforms often experience missing items, mistaken orders, and lost or missed deliveries. An increased reliance on robotic technology may reduce such errors, creating a better experience for customers and merchants alike.

•Safer Roads: According to a report by the Governors Highway Safety Administration (the “GHSA”) published in July 2025, there were 7,148 pedestrians killed in motor vehicle crashes in 2024 in the United States and 1,166 cyclists killed in motor vehicle crashes in 2023. There are many factors that contribute to pedestrian and bicyclist fatalities and injuries as a result of motor vehicle accidents, including: speeding, distracted driving, drunk driving, and aggressive driving. Sidewalk delivery robots weigh significantly less and move at lower speeds than cars. We believe that replacing cars with sidewalk delivery robots could result in safer cities for pedestrians and cyclists.

•Improved Healthcare Efficiency and Patient Care: In healthcare settings, we believe robotic automation can address critical labor shortages by performing routine logistical tasks such as transporting supplies, medications, lab samples, linens, and equipment. Nurses may spend a significant portion of their shifts on non-clinical tasks, including walking to retrieve supplies and making deliveries. By allowing clinical staff to focus on direct patient care rather than non-clinical logistics, we believe autonomous robots like Moxi can improve both operational efficiency and patient outcomes.

Technology

Our robotic technology has been developed based on the following key principles:

1.Humans and Machines: We believe that the most effective way to quickly realize the benefits of AI and autonomy in our lives is by designing solutions that leverage both human and machine intelligence in ways that are collaborative and complementary. Creating fully autonomous machines that are safe and reliable without any human intervention requires substantially more time and capital investment than creating machines that are mostly automated but can rely on occasional human support, especially when it comes to high consequence safety-critical decisions.

2.Labor-Optimized: We believe the cost of advanced sensors and hardware will continue to decrease over the coming years, and optimizing against such costs can be premature and may yield diminishing long-term

returns. Instead, we have continued to innovate by designing highly capable hardware and software solutions that optimize against the largest cost of delivery and logistics: labor.

3.Robots Among People: We believe that building world-class hardware, software, AI and autonomy for robots to share spaces with people positions us to build market value and create a lasting legacy.

4.Robotic Platform: Our mission is to build a superior robotic platform that can accelerate the adoption of robotic solutions beyond last-mile delivery.

The following are key highlights of our approach to technology development:

Artificial Intelligence

AI empowers Serve robots to navigate outdoor and indoor environments and interact with their environment safely and efficiently. We use the latest AI methodologies to design, train and deploy a host of models on Serve robots. Currently, our AI models are used to perform a variety of tasks, including identification of sidewalk surfaces, intersections, doorways, elevators, traffic signals, obstacles, pedestrians and vehicles, and projecting the trajectory of other dynamic agents.

Since we began development, we have continued to create AI models with new capabilities, while improving the performance of existing models. We expect to continue developing increasingly capable AI models to improve our robots’ performance and differentiate Serve from alternative products and solutions. In addition, we have completed various acquisitions of businesses engaged in the development of AI models with novel and differentiated capabilities, with the intention of further expanding and enhancing the Company’s portfolio of AI technologies, accelerating innovation, and supporting the integration of advanced capabilities across its products and services. As the broader field of AI advances, we expect to benefit from such advances by increasing the efficiency and effectiveness of our robots.

Level 4 Autonomy

In January 2022, we announced the deployment of a new generation of sidewalk delivery robots capable of operating at Level 4 autonomy. Level 4 autonomous robots can operate without humans in the loop for periods of time while they are operating in their intended operating environments (also known as “ODDs”). Specifically, our robots are capable of driving autonomously in certain environments without a remote human supervisor having to oversee their movement. This capability makes it possible to operate our robots at lower cost than remotely operated robots used by our competitors, because it enables a single remote supervisor to oversee multiple deliveries simultaneously.

Through frequent software updates, Serve’s AI models are continuously improved. We believe that over the coming years, we can steadily increase our robots’ autonomous capabilities using new and improved AI models and more training data including new edge cases encountered every day by our operating fleet.

Safety

Achieving Level 4 autonomy requires our robots to uphold rigorous safety standards. Organizationally, we adhere to the four standard components of a Safety Management System: Safety Risk Management, Safety Assurance, Safety Promotion, and Safety Policy. Together these components support a high level of safety performance and enable continuous safety improvement.

Operationally, each robot is equipped with robust onboard safety systems powered by Serve’s advanced sensors and AI capabilities. While highly trained remote supervisors are available to assist in specific, predefined situations, the robot is designed to operate safely and independently. This layered approach significantly reduces the likelihood of collisions, including during supervised operation.

At Serve, we believe that robots that rely primarily on human oversight cannot achieve the highest levels of safety due to the potential for human error. As a result, we focus on advancing our technology through multiple layers of redundant onboard safety systems to enable safe and reliable operation without undue dependence on human intervention. These systems incorporate diverse sensor modalities, including active sensors (such as LiDAR and ultrasonics) and passive sensors (such as cameras), allowing our robots to navigate complex environments with a high degree of reliability.

Serve also applies best-practice processes for the development of safety-critical systems, ensuring these processes are scalable across product design, manufacturing, and operations. Clearly defined requirements and verification methods allow us to deliver products and services that are demonstrably safe.

Our robots have a wide range of capabilities related to safety, such as:

•Automatic emergency braking: If a remote supervisor unintentionally places a robot at risk of collision, the robot is designed to automatically override the command and stop.

•Obstacle detection and avoidance: Our robots use multiple sensor modalities to detect and avoid obstacles in real time, including vehicles, pedestrians, furniture, medical equipment, and other dynamic objects in both outdoor and indoor environments.

•Fail-safe mechanical braking: In the event of a power, electronic, or compute failure, robots are designed to automatically come to a complete stop using a mechanically designed fail-safe braking system. This provides an additional layer of protection beyond electronic braking alone.

Manufacturing

Our robots are designed by Serve’s internal team of mechanical, electrical, and systems engineers to meet product requirements while optimizing for component availability and manufacturing scalability.

Most components are widely commercially available or fabricated from raw materials using common manufacturing processes including machining, molding, stamping, and additive manufacturing from multiple sources. However, certain highly complex components are obtained from single or limited sources, and we may have to compete for access to such components against other participants in the robotics, consumer electronics, and automotive markets. These components can at times be subject to industry-wide shortages, resulting in long lead times and significant pricing fluctuations. Under these circumstances and to maintain production schedules, it may be necessary to temporarily source alternate higher-priced compatible components. Key suppliers of single and limited source components include NVIDIA Corporation, Ouster, Inc., and other vendors providing cameras, ultrasonic sensors, electronic motors, and modems. To mitigate such supply risks, we routinely search for more available alternatives and/or enter into strategic partnerships and agreements to secure pricing and supply of components.

Manufacturing and assembly of robots are performed by third-party contract manufacturers. We also engage a contract manufacturing firm for the production of various subassembly components.

Business Strategy

After more than eight years of research and development investment in AI, autonomy, safety, and efficiency of our robots, we are in a leading position to partner with some of the world’s largest food delivery platforms, restaurants, retailers, healthcare systems, and convenience brands to augment their logistics and delivery capabilities with autonomous robotics solutions.

Through a series of commercial agreements with delivery partners, the Company has expanded its operational footprint and achieved a significant deployment milestone, with approximately 2,000 sidewalk delivery robots deployed across multiple geographic markets. These contractual relationships enable the Company to integrate its autonomous delivery solutions into partner platforms and support scaled commercial operations. We plan to continue growing our delivery operations and establish Serve as a global leader in automated last-mile delivery. We also intend to extend the applications of our autonomy platform beyond sidewalk delivery into complementary environments and verticals where autonomous navigation among people is required, such as indoor logistics and healthcare settings. Our growth will be facilitated by continued investment in our hardware, software and AI developments that increase the performance and efficiency of our fleet. Looking ahead, we plan to increase utilization of our robot fleet by advancing performance, safety, and user experience; broadening use cases and partnerships; and expanding integration with third-party platforms and services.

Our robots can perform other value-add tasks while in operation and capture additional revenues, such as out-of-home (“OOH”) branding. Similar to billboards and buses, brands are placing ads on Serve robots’ exteriors. Our robots are also utilized in experiential advertising initiatives, including one-time promotional events, brand activations, and appearances in film and other media. These engagements allow advertisers to leverage the visibility and novelty of our robotic platforms to enhance brand awareness and consumer engagement in real-world environments.

We are also engaging partners in software-based services that complement our core robotic operations, including recurring software licensing and data monetization. Alongside our strategic partners, we continue evaluating additional applications of the technology we have developed.

Competition

The global market for autonomous robots is highly competitive, and we expect competitive conditions to intensify as a significant and increasing number of established companies and new entrants seek to enter or expand within this market. In addition, we face competition from traditional methods, including human couriers in last-mile delivery and manual logistics workflows in healthcare, as well as from other autonomous robotics providers.

Environmental Impact

According to the International Energy Agency, passenger vehicles emitted 3.2 billion metric tons of CO2 in 2023. The Bureau of Transportation Statistics estimates that in 2023, 16.2% of car trips were taken for shopping and errands, which are activities that are particularly well-suited to robotic delivery. If half of all shopping trips could be completed by sidewalk delivery robots, the scaled deployment of robotic delivery could reduce global passenger car emissions attributable to such trips from approximately 259 megatons to less than 10 megatons per year. This is a reduction of over 96% as estimated by a study published in Transportation Research Part D: Transport and Environment (Volume 85, August 2020, 102443).

Government Regulations

In the United States, delivery robots are generally allowed to operate on most sidewalks by default. Over twenty-five states and a number of cities have put in place legal frameworks to explicitly permit the use of sidewalk delivery robots. The instances of cities welcoming sidewalk delivery robots exceed the examples in which their operations were banned. This positive regulatory momentum for sidewalk delivery robots may be due to a number of factors:

•Robots Inherently Safer: Cars carry more kinetic energy than sidewalk delivery robots and therefore pose a significantly greater risk of serious injury or fatality. In the United States alone, vehicle collisions result in approximately 40,000 deaths per year and injure thousands of additional pedestrians and cyclists. Replacing unnecessary car trips with robotic delivery can make cities safer for pedestrians and cyclists.

•Robots Reduce Emissions: Many cities have established targets for reducing carbon emissions, and robots can help advance these environmental initiatives by offering a more sustainable delivery option that produces a fraction of the emissions associated with personal vehicles.

•Robots Reduce Congestion: Deliveries often originate in high-traffic areas where restaurants and shops tend to be located. Shifting these trips from cars to robots can reduce traffic congestion and ease parking challenges in urban areas.

•Robots Reduce Delivery Cost for Local Merchants: Recently, a number of cities have implemented restrictions on how much delivery platforms can charge restaurants for their services. Robots can offer a lower-cost alternative, helping merchants remain competitive while lowering their overall delivery expenses.

We are subject to various federal, state and local laws and regulations that govern aspects of our business operations. While costs associated with compliance with laws and regulations have increased as the number and scope of regulations have increased, the total costs incurred have not had, and are not currently expected to have, a material effect on our capital expenditures, results of operations or competitive position. See Part I, Item 1A. “Risk Factors” for discussion of risks relating to federal, state, local and international laws and regulations applicable to our business.

Intellectual Property

Our success and ability to compete are significantly dependent on our core technology and intellectual property. As of December 31, 2025, we had 62 active patent matters in various jurisdictions: Australia (1 patent), Canada (6 patents), China (2 patents), the European Patent Office (3 patents), Japan (2 patents), the Republic of Korea (2 patents), the United States (44 patents) and the World Intellectual Property Organization (2 patents). As of December 31, 2025, we have been granted 36 of the 62 patents for which we have applied — 4 in Canada, 1 in China, 2 in the European Patent Office and 29 in the United States. Our granted patents will expire between May 2029 and June 2042, assuming that all maintenance fees are paid, no portion of the patent has been terminally disclaimed and the patent has not been invalidated. In certain jurisdictions, and in certain circumstances, patent terms can be extended or shortened.

We seek to protect our intellectual property and proprietary rights, including our proprietary technology and software, by relying on a combination of federal, state, and common law rights in the United States and other countries, as well as on contractual measures. However, these laws, agreements, and procedures provide only limited protection for our technology.

Although we currently primarily operate within the United States, the laws of some foreign countries do not protect our proprietary rights to as great an extent as domestic laws, and many foreign countries do not enforce these laws as diligently as government agencies and private parties in the United States.

We encourage our employees to continue to invent and develop new technologies so as to maintain our competitiveness in the marketplace. We generally enter into confidentiality and non-disclosure agreements with our employees, consultants, vendors and customers, and generally limit access to and distribution of our proprietary information. We also enter into invention assignment agreements with our employees and consultants, which give us the rights to their inventions during the term of the agreements.

Despite our best efforts, our means of protecting our proprietary rights may be inadequate. It is possible that our current patents, or patents we may later acquire, may be successfully challenged or invalidated in whole or in part. It is also possible that we may not obtain issued patents for our pending patent applications or other inventions we seek to protect. In that regard, we sometimes permit certain intellectual property to lapse or go abandoned under appropriate circumstances and due to uncertainties inherent in prosecuting patent applications, sometimes patent applications are rejected and we subsequently abandon them. It is also possible that any patent issued to us may not provide us with any competitive advantages, or that the patents of others may harm or altogether preclude our ability to do business. Moreover, unauthorized parties and our competitors may attempt to copy aspects of our products or obtain and use information that we regard as proprietary, or independently develop technology that is similar to ours thereby rendering our protected products less valuable if the design is favorably received in the marketplace.

Human Capital Management

We strongly believe our employees are our greatest assets. We value and support hiring exceptional talent to develop our core technology and drive our business growth. We strive to meet these objectives by offering competitive pay and benefits in a diverse, inclusive, and safe workplace. In addition, we provide opportunities for our employees to grow and develop their careers.

As of December 31, 2025, we employed a total of 370 full-time and 10 part-time employees globally, of which 359 employees were based in the United States, and 21 employees were based in Canada through our wholly-owned Canadian subsidiary. By primary job function, approximately 38% of our employees have engineering or product roles, 49% are in operations, and 13% have business development or other administrative roles. None of our employees are represented by a labor union or covered by a collective bargaining agreement. We consider our relationship with our employees to be in good standing.

Competitive Pay and Benefits

We provide compensation and benefits packages that we believe are competitive within our industry. We use a combination of cash and equity compensation, and other benefits to attract, motivate and retain our employees, including equity awards, retirement programs, flexible or paid time off based on department, and health and wellness benefits. In addition, we benchmark our compensation and benefits packages periodically to remain competitive with our peers and attract and support our talent acquisition and retention efforts.

Employee Recruitment, Retention, and Development

We believe our corporate culture, competitive compensation and benefits programs, and career growth and development opportunities promote employee tenure and retention and reduce employee turnover. We have maintained high employee retention since becoming an independent company in 2021 and continue to monitor employee turnover rates closely, as our success depends on retaining and investing in our highly skilled technical staff.

Diversity, Equity, and Inclusion

We are committed to fostering a diverse, equitable, and inclusive workplace. We believe that an inclusive environment, that recognizes and values the unique backgrounds, experiences, and perspectives of our employees, enhances our organizational effectiveness. Accordingly, we seek to cultivate a culture in which all employees are empowered to contribute meaningfully to the Company’s success.

Safety, Health, and Wellness

We prioritize the safety, health, and well-being of our employees. We are committed to an injury-free workplace and provide comprehensive workplace training and support to reduce or eliminate health and safety risks.

Available Information

We are a public company, and our Annual Reports on Form 10-K, Quarterly Reports on Form 10-Q, Current Reports on Form 8-K and any amendments to such reports are filed with the Securities and Exchange Commission (the “SEC"). We are subject to the informational requirements of the Exchange Act and file or furnish reports, proxy statements, and other information with the SEC. Such reports and other information filed by us with the SEC will be available free of charge on our website at www.serverobotics.com when such reports are available on the SEC’s website. The SEC maintains a website that contains reports, proxy and information statements, and other information that issuers file electronically with the SEC at www.sec.gov. Our website is www.serverobotics.com, and we can be contacted at investor.relations@serverobotics.com.

The contents of the websites referred to above are not incorporated into this filing. Further, our references to the URLs for these websites are intended to be inactive textual references only.