In a development that underscores the accelerating convergence of artificial intelligence and physical robotics, Google has taken a direct financial stake in Apptronik, a Texas-based developer of humanoid robots. The investment arrived as part of a substantial Series A funding round that the company announced in February 2025, with the round ultimately closing at a total of $415 million.
The funding round was initially structured at $350 million, with B-Capital and Capital Factory serving as co-leads. Google participated as an investor in this tranche. According to reporting from CNBC, the round was oversubscribed, meaning demand from investors exceeded the initially targeted amount. As a result of this excess interest and additional investment that followed, the Series A total was raised to $415 million. This distinction between the initially announced $350 million figure and the final $415 million total is an important detail for observers tracking the company's financial trajectory, as it indicates strong market appetite for humanoid robotics ventures.
Apptronik's chief executive, Jeff Cardenas, confirmed the funding details in an interview with CNBC's "Squawk Box" program, noting that the round included backing from Google. The company had previously raised $28 million in earlier funding rounds, according to the same reporting. This new injection of capital represents a significant step up in scale for the company, which was founded in 2016.
The funding is earmarked for several specific purposes. Apptronik has stated that the capital will help it expand the deployment of its humanoid robot, Apollo, and accelerate the development of that platform. Additionally, the company plans to grow its team and scale its operations more broadly. Cardenas has also indicated that the funding will support the company's artificial intelligence robotics development efforts, with a long-term ambition to eventually sell robots at a price point comparable to that of an automobile. That pricing goal, while ambitious, provides a useful reference point for potential buyers trying to understand where the market for humanoid robots is headed.
Beyond the financial transaction, Google's involvement extends into the technical domain. In March 2025, Google's DeepMind division announced a partnership with Apptronik focused on building what the two companies describe as "the next generation of humanoid robots." This collaboration will leverage Gemini 2.0, Google's generative AI large language model. Google has described Gemini 2.0 as its "most capable" AI to date. The partnership was announced via a Google blog post, and the companies have demonstrated some early results in video form.
Those demonstration videos, which were shown as part of Google's announcement of two new AI models for robotics, depict Apptronik robots performing tasks such as plugging items into power strips, filling a lunchbox, and moving plastic vegetables. These tasks, while seemingly simple for a human, represent meaningful milestones for a humanoid robot operating with AI-driven autonomy. Both of the new AI models announced by Google DeepMind run on Gemini 2.0.
One of those models, called Gemini Robotics-ER, is designed specifically for roboticists to use as a foundation for training their own models. Google has made this model available to Apptronik as well as to a group of "trusted testers" that includes Agile Robots, Agility Robots, Boston Dynamics, and Enchanted Tools. This distribution strategy suggests that Google is positioning itself not merely as a partner to a single robot maker but as a platform provider for the broader robotics industry.
Why it matters for European robot service
For readers of Robot Service Map who track the European robotics landscape, this development carries several implications that extend well beyond the borders of Texas or the balance sheets of Silicon Valley.
First, the scale of the funding round itself is notable. A $415 million Series A for a humanoid robotics company signals that institutional investors see a clear path to commercial viability for general-purpose humanoid robots. When Google, a company with near-unlimited resources and its own AI ambitions, chooses to invest in a hardware company rather than build its own humanoid platform from scratch, it suggests that the hardware side of humanoid robotics is seen as a domain where specialised expertise matters. For European companies working in this space, this is a signal that the competitive bar is rising, and that capital is flowing to the leaders.
Second, the Google DeepMind partnership has implications for how AI and robotics will be integrated in the coming years. Gemini 2.0 is a large language model, a type of AI that has primarily been associated with text generation and conversational interfaces. Applying such a model to physical robot control is a significant technical step. The demonstration videos showing Apollo robots performing manipulation tasks under AI guidance offer a glimpse of what service robots might be capable of in the near term. For European robot service providers and integrators, this suggests that the AI layer of robotics is becoming more powerful and more accessible, potentially lowering the barrier to deploying capable automation in logistics, manufacturing, and other service environments.
Third, the involvement of Google as both an investor and a technology partner raises strategic questions for the European robotics ecosystem. Europe has its own strengths in robotics, with companies like Agile Robots and Enchanted Tools appearing on Google's list of trusted testers for Gemini Robotics-ER. The fact that European companies are included in this early access group is a positive sign, indicating that Google sees value in engaging with the European ecosystem rather than bypassing it. However, it also means that European robot makers are increasingly likely to build their AI capabilities on top of American foundation models. This dependency carries both benefits and risks. The benefits include access to state-of-the-art AI capabilities without the need for massive in-house AI research budgets. The risks include potential supply-chain vulnerabilities and the possibility that strategic decisions made in Mountain View could affect the roadmap of European robot deployments.
Fourth, the funding round and partnership come at a time when the humanoid robot market is attracting intense attention globally. Tesla's Optimus robot is frequently cited as a competitor, and CNBC's reporting explicitly frames Apptronik as a rival to Tesla in this domain. The competitive dynamics between these American companies will inevitably shape the global market, including Europe. European buyers and operators who are considering humanoid robots will have more options than ever before, but they will also need to navigate a landscape where the underlying AI technology is concentrated in the hands of a few large players.
Finally, the involvement of NASA and Nvidia in Apptronik's development history is worth noting. Apptronik has worked with NASA on the humanoid robot Valkyrie, and has secured partnerships with Nvidia as well. These relationships lend credibility to the company's technical approach and suggest that its robots are being developed with input from some of the most demanding technical organisations in the world. For European buyers, this track record may be reassuring, as it indicates that the technology has been vetted by institutions with rigorous standards.
What buyers and operators should know
For organisations in Europe that are evaluating humanoid robots for deployment in manufacturing, logistics, or other service environments, the Apptronik news offers several points of consideration.
The company's commercial strategy is already taking shape. Apptronik has struck deals with Mercedes-Benz and GXO Logistics to pilot Apollo at manufacturing and warehouse facilities. These are not speculative partnerships; they are real deployments with major commercial entities. Mercedes-Benz is one of the world's most recognisable automotive manufacturers, and GXO Logistics is a major player in the warehousing and logistics sector. The fact that these companies have chosen to pilot Apollo suggests that the robot has reached a level of maturity that makes it worth testing in real operational environments. For European operators, this is a useful data point, as it indicates that humanoid robots are moving from the laboratory to the factory floor.
Apollo itself was first released as a commercial product in 2023. It is described as a general-purpose humanoid robot, meaning it is designed to handle a wide range of tasks rather than being specialised for a single function. The robot mirrors the look of a human, a design choice that is deliberate. By mimicking human form, the robot can use the same equipment and work in the same spaces as humans, without requiring significant modifications to existing facilities. This is a key selling point for humanoid robots generally: they promise to fit into environments that were designed for people, rather than requiring the environment to be redesigned around the machine.
Apptronik's history predates Apollo. The company was founded in 2016 and has developed bipedal mobility platforms, upper body humanoid robots, and wearable robotic systems. This breadth of experience means that Apollo is not a first attempt but rather the culmination of years of iterative development. The company's work on NASA's Valkyrie robot is particularly relevant, as it demonstrates an ability to work on advanced, government-funded robotics projects with demanding technical requirements.
The pricing ambition articulated by Cardenas is worth examining. The goal of eventually selling robots for the price of a car is a useful benchmark, but it is important to note that this is an aspiration rather than a current reality. The source material does not disclose current pricing for Apollo, nor does it provide any specific figures for total cost of ownership, maintenance, or service contracts. Buyers should therefore treat the "price of a car" comment as a directional signal rather than a concrete commitment. It suggests that the company is aiming for a price point that would make humanoid robots accessible to a broad range of businesses, but the actual pricing will depend on many factors, including production scale, component costs, and market competition.
The source material also does not disclose specific technical specifications for Apollo, such as payload capacity, battery life, or operational uptime. It does not provide details on service-level agreements, response times, or spare-part lead times. These are critical considerations for any organisation planning to deploy robots in a production environment, and the absence of such information in the public record means that buyers will need to engage directly with Apptronik or its partners to obtain these details. Robot Service Map advises operators to ask pointed questions about these topics before making any commitments.
The Google partnership adds another layer of consideration. The integration of Gemini 2.0 into Apptronik's robots means that the AI capabilities of these machines will be closely tied to Google's technology roadmap. For buyers, this is generally positive, as Google has substantial resources and a track record of advancing AI capabilities. However, it also means that the robots' intelligence will depend on a third-party platform. Organisations with strict data governance requirements should investigate how the AI models are deployed, whether they run on-device or in the cloud, and what data is transmitted to Google's infrastructure. The source material does not address these questions, so they remain open items for due diligence.
The competitive landscape is also worth monitoring. Google is not the only major technology company pursuing AI for robotics, and the source material notes that Google is "far from alone" in this pursuit. The availability of Gemini Robotics-ER to multiple robot manufacturers, including Boston Dynamics and Agility Robots, suggests that the AI layer of robotics is becoming a shared resource rather than a proprietary advantage for any single hardware maker. This could accelerate the pace of innovation across the industry, but it also means that the differentiation between robot manufacturers will increasingly come down to hardware quality, software integration, and service support rather than AI capability alone.
For European buyers, the key takeaway is that the humanoid robot market is maturing rapidly, and the involvement of major technology companies like Google is a sign that this is not a niche interest but a mainstream industrial trend. The funding round, the partnerships with Mercedes-Benz and GXO Logistics, and the technical collaboration with Google DeepMind all point to a company that is serious about commercial deployment. At the same time, the absence of detailed pricing, service, and specification data in the public record means that buyers must conduct thorough due diligence before making procurement decisions. The technology is promising, but the commercial details remain to be clarified on a case-by-case basis.
Sources
https://www.cnbc.com/2025/02/13/tesla-humanoid-robots-rival-apptronik-350-million-funding-round-google.html
Published by Vigla Media OÜ (Estonia).