Meta Tests Cable-Swapping Robots That Could Automate Data Center Work

Meta Tests Cable-Swapping Robots That Could Automate Data Center Work

Meta is testing robots inside its data centers to handle maintenance jobs including swapping network cables, power-cycling servers and reseating hardware, using systems from Watney Robotics, Kinova and ABB. The trials are extending the company’s automation efforts beyond logistics and inventory work into tasks still largely performed by technicians.

The experiments remain supervised and, in several cases, slower or more limited than human workers. Current and former employees familiar with the projects described charging downtime, navigation problems and difficulties handling the dense cabling and hardware layouts found inside data centers.

At Meta’s Altoona, Iowa, facility, the company has been testing two dual-armed Watney robots on cabling work since June 2025. Human operators still oversee the machines, which have not yet matched technicians for speed. Workers also cited the need to recharge the robots as another constraint on their usefulness.

Cabling is one of the more difficult areas Meta is trying to automate. The company has acknowledged internally that its current robots cannot handle some of the more demanding cable work required for systems built around NVIDIA GB300 hardware. A former employee said data-center equipment and repair procedures were designed around people, making it harder to adapt existing infrastructure for robotic handling. Meta is also evaluating a Kinova Gen3 robotic arm for server power cycling. At some facilities, simpler robotic equipment can remotely press the power button on machines such as Mac Minis after receiving a human command.

At the company’s Prometheus campus in New Albany, Ohio, ABB robots are being tested for hardware reseating. Those machines use a wheeled base, a lift mechanism and a six-axis robotic arm, with additional maintenance tasks and lower levels of human oversight among the possible future goals. Those projects are less mature than Meta’s existing automation systems. The company already operates self-driving tugger robots that move heavy server racks and an internally developed wheeled robot that scans equipment and tracks inventory across multiple data centers, including sites in Iowa and Virginia.

Even those established systems still depend on people for certain jobs. Meta’s inventory robot uses a grayscale camera, preventing it from distinguishing some green and red equipment lights. Workers said it can also struggle with corners and cables on the floor. People must open doors for the machines and remotely move them between buildings.

The potential labor impact is one reason the cable trials are drawing attention. One Meta data-center worker estimated that a successful cable-swapping robot could eliminate as much as 80% of some technicians’ workloads. That figure reflects the worker’s estimate rather than a company projection.

Meta has taken a different public position on its workforce needs. The company says it faces a shortage of skilled infrastructure workers and is investing in hiring and training. It has also launched a program intended to train thousands of people each year in electrical, mechanical and plumbing trades.

Eric Xu, a senior manager for robotics at Meta, said at a conference last year that the company ultimately wants robots to help data centers respond to incidents faster, monitor operating conditions and perform preventive maintenance.

Other technology companies are pursuing similar ideas. Microsoft and Google have invested in data-center robotics, while Amazon has discussed using robots for recycling components.

For Meta, however, the current trials show that automating physical repair remains more difficult than automating movement or inspection. Robots can already transport racks and scan inventory, but the maintenance systems being tested for cabling, server resets and hardware reseating still require supervision and face practical limits that technicians do not.

This analysis is based on reporting from superpowerdaily.

Image courtesy of Meta.

This article was generated with AI assistance and reviewed for accuracy and quality.

Updated Aug 28, 2026

About this article: This article was generated with AI assistance and reviewed by our editorial team to ensure it follows our editorial standards for accuracy and independence. We maintain strict fact-checking protocols and cite all sources.

Word count: 610Reading time: 0 minutes

📧 Stay Updated

Get the latest AI news delivered to your inbox every morning.

AI News Daily

Breaking Intelligence • Since 2023

Join hundreds of thousands of AI professionals who start their day with our curated newsletter. Get breaking news, expert analysis, and exclusive insights.

Stay Ahead of AI

Get the latest AI breakthroughs, tools, and insights delivered to your inbox every week.

Free forever Unsubscribe anytime No spam guarantee

Go Premium

Unlock unlimited AI tools and an ad-free reading experience designed for AI professionals.

• Ad-free experience• Premium AI tools
Start Free Trial

14-day free trial • Cancel anytime
Plus $9/mo • Pro $90/yr (2 months free)

Follow Our Community

ChatAI

Breaking Intelligence

Your daily briefing on what matters in AI. Trusted by developers, researchers, executives, and AI enthusiasts worldwide.

© 2026 ChatAI. All rights reserved.