Yazılar

Nvidia and Foxconn in Talks to Deploy Humanoid Robots at New Houston AI Server Factory

Taiwanese electronics manufacturer Foxconn and U.S. AI chipmaker Nvidia are reportedly negotiating to introduce humanoid robots at Foxconn’s upcoming factory in Houston, which will produce Nvidia AI servers. According to sources familiar with the discussions, this would mark the first time Nvidia products are manufactured with the help of humanoid robots and Foxconn’s inaugural use of such robots in an AI server production line.

The planned deployment, expected to be finalized within months, represents a significant advancement in the use of human-like robots in manufacturing, potentially transforming factory processes. Foxconn is working on its own humanoid robots in collaboration with Nvidia and has also tested humanoids developed by China’s UBTech. Details on the specific types, appearances, and number of robots planned for the Houston facility remain unclear.

The goal is to have the humanoid robots operational by the first quarter of next year, coinciding with the start of production for Nvidia’s GB300 AI servers at the Houston plant. While precise tasks have not been confirmed, Foxconn has been training humanoid robots for activities such as picking and placing objects, cable insertion, and assembly, according to a May company presentation.

Foxconn’s Houston factory is particularly suited for humanoid robot deployment due to its newness and ample space compared to existing AI server production sites. Nvidia and Foxconn declined to comment on the matter, and sources spoke anonymously due to lack of authorization to speak publicly.

At a recent event in Taipei, Leo Guo, general manager of Foxconn Industrial Internet’s robotics unit, revealed plans to showcase two humanoid robot models at Foxconn’s annual tech event in November—one with legs and another on a wheeled autonomous mobile robot base, the latter being the more cost-effective option.

Nvidia announced in April its plans to build AI supercomputer manufacturing plants in Texas, partnering with Foxconn in Houston and Wistron in Dallas, with production ramp-up expected within 12 to 15 months.

For Nvidia, integrating humanoid robots in AI server manufacturing signifies a deeper commitment to robotics technology, building on its existing platform that supports humanoid robot development. Nvidia CEO Jensen Huang has forecasted that widespread use of humanoid robots in manufacturing is less than five years away.

Several automakers, including Mercedes-Benz and BMW, have experimented with humanoid robots on production lines, while Tesla is developing its own. China also heavily supports humanoid robotics, anticipating that many factory tasks will eventually be carried out by these robots.

UAE’s ‘Stargate’ AI Megaproject to Launch in 2026 with 100,000 Nvidia Chips

A landmark AI infrastructure project in Abu Dhabi, dubbed Stargate UAE, is set to begin operations in 2026, becoming one of the largest and most advanced artificial intelligence data centers outside the United States. The project will be powered by an estimated 100,000 Nvidia chips, marking a significant step in the UAE’s ambition to become a global AI superhub.

The 1-gigawatt first phase is part of a broader 5-gigawatt, 10-square-mile site and represents the first outcome of a U.S.-brokered deal led by President Donald Trump. The partnership involves a collaboration between UAE’s state-backed G42, and major U.S. and Japanese tech firms: OpenAI, Oracle, Nvidia, Cisco Systems, and SoftBank.

Key Highlights:

  • Launch Timeline:

    • First phase (200MW) to be operational by 2026

  • Hardware:

    • Uses Nvidia Grace Blackwell GB300 systems, Nvidia’s most powerful AI chips

    • TrendForce estimates point to 1,400 servers, each with 72 chips, totalling ~100,000 chips

  • Scale:

    • Part of a 5-gigawatt megaproject to house the largest AI infrastructure outside the U.S.

    • Located on a 26 sq-km site in Abu Dhabi

  • Purpose:

    • Will allow UAE government bodies and businesses to access cutting-edge generative AI models

    • Positioned as a “first-in-the-world platform”, according to Oracle’s Larry Ellison

U.S. Policy Reversal

The announcement follows a significant policy shift by the Trump administration, which reversed Biden-era restrictions on exporting advanced AI chips to the UAE due to concerns over the country’s ties to China.

The U.S. Commerce Department will oversee a bilateral working group with the UAE to ensure compliance with:

  • National security standards

  • Ethical AI deployment

  • Global oversight of AI infrastructure

Strategic Implications

  • Geopolitical Impact:

    • The project cements UAE’s growing alignment with U.S. tech interests, even amid global tensions over AI dominance

    • Could challenge China’s AI influence in the Middle East and Africa

  • Commercial Influence:

    • Strengthens partnerships between OpenAI, Nvidia, and Oracle, anchoring them in the Gulf region

    • Offers a new AI deployment and monetization model based on national infrastructure rather than global cloud platforms

  • Regional AI Race:

    • The Stargate initiative follows Saudi Arabia’s recent moves to create its own multimodal Arabic LLMs

    • Marks a regional AI arms race underpinned by state-backed funding and global tech partnerships