Meta Tests Robots in Data Centers That Could Replace 80% of Human Workload

Meta Deploys Robots in Data Centers for Critical Tasks

Meta is actively testing robots in data centers to perform tasks ranging from power cycling servers to swapping networking cables, according to multiple current and former workers familiar with the projects.1

1

2

2

The previously unreported initiative involves robotic arms and automated systems from vendors including Watney Robotics, Kinova, and ABB. This data center automation effort comes as Meta's AI infrastructure spending continues to soar, with the company seeking ways to manage operational costs across its rapidly expanding data center footprint.

In one experiment, Meta is evaluating whether a Kinova Gen3 robotic arm could handle power cycling or cutting off electricity to servers.1

1

The company is also testing a separate robot designed specifically for swapping networking cables. One Meta data center worker estimates that if successful, the cable-swapping bot could replace up to 80% of some people's workloads.2

2

At certain facilities, Meta has quietly adopted a simpler robot that resembles a finger or pointy stick, capable of pressing the power button on Mac Minis or other devices when remotely prompted by a human.3

3

Robots in Data Centers Address Technical Challenges

Eric Xu, senior manager for robotics at Meta, outlined the company's long-term goals at a conference last year, stating that robots in data centers could speed up incident response, conduct environmental monitoring, and handle preventative maintenance.1

1

"We believe more collaboration and research will be needed, but we have to start, otherwise we don't have a chance to do this," Xu said. The company has already demonstrated self-driving robots capable of moving heavy server racks, with deployments at its Iowa and Virginia facilities.3

3

Until recently, robots weren't well suited for the delicate work in data centers. They were expensive, and in earlier industry trials sometimes crushed servers while attempting simple tasks.2

2

However, hardware costs have decreased significantly, and AI models powering robots are now far more capable than a few years ago. Several startups are working to build commercial robots capable of assembling servers, swapping parts, and replacing wires. The growing complexity of AI data centers presents challenges for robots, making fine motor skills increasingly important for handling costly and vital components.3

3

Job Displacement Concerns Mount Among Workers

The automation push is generating anxiety among data center technicians. "We thought those of us performing the physical tasks were safe for a while, but not anymore," one worker told reporters. "It's coming for us all, unfortunately."1

1

In Altoona, Iowa, where Meta operates its largest data center campus employing hundreds of workers, group chats have taken on a demoralized tone, with employees "saying we're all going to be gone in a few years anyway because of the robots," according to a person familiar with the discussions.2

2

Workers also worry that even if robot adoption proves slower than anticipated, the quality of available jobs is trending downward. Two workers say Meta is developing AI software and other technologies that could allow it to hire lower-skilled, lower-paid workers and centralize roles in financially advantageous locations like Denver, Colorado.1

1

4

4

This shift could reduce demand for experienced technicians while moving remaining tasks to employees following AI-generated instructions.

Industry Trends and Future Implications

Robots could offer more consistent output at lower cost than human labor, especially in less populous regions where data centers are plentiful but qualified workers are scarce.2

2

Paul Golding, who oversees physical intelligence at chipmaker Analog Devices, notes that "more and more of our customers are telling us 'we want to run our data centers using humanoid [robots]' because you can run them at higher temperatures, in the dark, and all these things."1

1

Getting workers out of the picture could make it more feasible to run data centers underwater, in space, or other environments inhospitable to humans.

Nvidia and Alibaba are also developing systems to improve robot training. In June, Alibaba unveiled its Qwen-Robot Suite with models for navigation and object handling, while Nvidia researchers introduced AI agents that train robot fleets.4

4

ACE Robotics Chairman Wang Xiaogang predicts embodied AI could have a "ChatGPT moment" by the end of 2027, though limited real-world data remains a barrier.

Meta spokesperson Francis Brennan emphasized the company's continued investment in human workers: "America is in the middle of its biggest infrastructure boom since World War II, and there's a major shortage of skilled workers to fill the roles; we need more workers, not fewer."1

1

However, this message conflicts with worker experiences and the company's automation trajectory. The robots still require human supervision and struggle with obstacles, battery life, visual inspections, and dense cabling, but their development will determine whether Meta's expanding infrastructure creates jobs or replaces them.4

4