Google is making a major push into physical AI by partnering with robotics hardware companies to bring its Gemini AI models into the real world. The search giant's DeepMind division announced a partnership with Munich-based Agile Robots to integrate Gemini Robotics foundation models with Agile Robots' sensor-based robotic arms and humanoid robots. This move signals that Google sees robotics as one of the largest untapped use cases for artificial intelligence, competing directly against companies like Amazon and Tesla .

Why Is Google Suddenly Focused on Robotics?

Google's robotics strategy isn't random. The company recognizes that AI models trained in digital environments need real-world deployment data to improve. By partnering with hardware makers that already have thousands of robots in the field, Google gains access to valuable training data while helping its partners build smarter machines. Agile Robots alone has more than 20,000 deployed robotic systems globally, providing an immediate testing ground for Google's latest AI models .

The partnership will initially focus on high-value industrial use cases such as manufacturing tasks. This approach mirrors how other tech companies have approached AI development: start with controlled environments where the stakes are clear and the feedback loop is fast. Manufacturing offers exactly that kind of environment, where robots perform repetitive tasks and generate consistent data that can train better AI models.

"This research partnership is an important step in bringing the impact of AI to the real world. Agile Robots will help Google develop more advanced AI models for the next generation of robots," stated Carolina Parada, Senior Director and Head of Robotics at Google DeepMind.

Carolina Parada, Senior Director and Head of Robotics, Google DeepMind

How Is Google Building Its Robotics Ecosystem?

Google isn't relying on a single partnership. The company has adopted a multi-pronged strategy to establish itself as a central player in the robotics industry:

• Hardware Partnerships: Google is working with multiple robotics hardware makers including Agile Robots, Apptronik (a Texas-based developer), and Boston Dynamics (owned by Hyundai) to integrate its Gemini AI models across different robot platforms.
• Software Infrastructure: Google moved Intrinsic, a robotics software company, from its "Other Bets" category into the main company, positioning it as "The Android of robotics" with a focus on manufacturing and integration with Google's Gemini AI and data center infrastructure.
• Talent Acquisition: Google hired Aaron Saunders, the former Chief Technology Officer of Boston Dynamics, signaling serious commitment to building deep robotics expertise within DeepMind.

This ecosystem approach mirrors how Google built its dominance in cloud computing and mobile operating systems. Rather than trying to manufacture robots itself, Google is becoming the AI brain that powers robots built by others. The company debuted two new AI models in mid-2025: Gemini Robotics and Gemini Robotics-ER (extended reasoning), which translate generative AI into physical action commands that control robots .

What Does This Mean for the Broader Physical AI Market?

Google's aggressive robotics push reflects a broader industry trend. Robotics is surfacing as a key use case for AI across the entire tech sector. In February, Bedrock Robotics, an autonomous vehicle technology startup for construction machinery founded by veterans of Waymo and Segment, raised $270 million in funding, valuing the two-year-old company at $1.75 billion. The funding round was led by Alphabet's investment arm CapitalG, Valor Atreides A.I. Fund, and Nvidia's venture arm .

Meanwhile, Figure AI, a lesser-known robotics company backed by Nvidia, gained significant visibility when its third-generation humanoid robot appeared at a White House event alongside First Lady Melania Trump. The robot greeted attendees in multiple languages and described itself as "a humanoid built in the United States of America." Figure AI raised more than $1 billion in its Series C funding round in September, giving it a post-money valuation of $39 billion .

The White House spotlight on Figure AI underscores how robotics has become a national priority amid global tech competition. Beijing has also promoted humanoid robots at highly publicized events this year, signaling that both the U.S. and China view physical AI as a critical frontier for technological leadership .

What Challenges Remain in Physical AI Development?

Despite the excitement, the robotics industry faces significant hurdles. Figure AI's experience illustrates some of the safety concerns that persist. In November, the company was sued by its former head of product safety, Robert Gruendel, who alleged he was fired after warning executives that the company's robots were powerful enough to fracture a human skull. Gruendel claimed that Figure AI's next-generation robots moved at superhuman speed and generated force approximately twice the level necessary to fracture an adult human skull. He also alleged that one robot had carved a gash into a steel refrigerator door during a malfunction .

Figure AI contends that Gruendel was fired for poor performance and described the allegations as "falsehoods." The company countersued in January, saying Gruendel failed in his role to help build a safe robot. The lawsuit remains pending and has drawn attention to broader questions about safety standards in humanoid robotics development .

Google's robotics expansion has also sparked internal concerns. Some DeepMind employees raised concerns at an all-hands meeting earlier this year about Boston Dynamics' long-standing contracts with the Defense Department, according to reporting from Business Insider .

What's the Practical Impact for Businesses and Consumers?

For manufacturing companies, Google's robotics partnerships mean access to AI-powered robots that can learn and adapt to new tasks more quickly than previous generations. Agile Robots will integrate Google's technology into existing industrial robots at scale, starting with high-value manufacturing tasks. This could accelerate automation in factories and warehouses, potentially reducing labor costs while increasing productivity .

For consumers, the implications are longer-term. Figure AI has stated that its third-generation humanoids are applicable for commercial and household tasks, and the company aims to deploy thousands of robots in homes and logistics over the coming years. However, safety concerns and regulatory questions remain unresolved .

The broader takeaway is clear: major tech companies are treating physical AI as a defining frontier for the next decade. Google's multi-partnership strategy, combined with aggressive hiring and infrastructure investments, suggests the company is betting that whoever controls the AI models powering robots will shape the future of automation across industries. The race is on, and the stakes are high.