EXPLORINEER
en

DE — heise online

Humanoid Robots: Will AI Make Them Practical for Everyday Use Next Year?

Robotics1 min read

This is a preview of a heise+ article, a paid tech magazine. It asks whether humanoid robots (robots shaped like people) will be ready for everyday use next year, thanks to AI. The article comes from a tech conference in Berlin where Nvidia showed new tools.

What happened

At Nvidia's GTC 2026 conference in Berlin in October, the US chip company showed its newest high-power computers. It also announced an upgraded version of Isaac, its platform for building AI for robots in the physical world. The author went to the event to see which AI-controlled humanoid robots will be ready next year and how far they have come. Only a preview is free; the full article is for paying subscribers.

Smarter robots

Humanoid robots are getting smarter. New models use stronger AI processors. They learn from the experience and working routines of other robots. They are trained to learn one skill after another, and the number of skills they know keeps growing.

Why it matters

Humanoid robots could soon work in factories and help with services. Today they can already handle existing spaces, such as a workshop or a home. They can climb a few steps or cross a ledge. Because their AI lets them react on their own to changing situations, they could become useful in everyday life.

What to watch

  • Which new humanoid robots are ready for real use next year
  • Whether robots can handle many different skills in factories and homes
  • How well the upgraded Isaac platform helps developers build robot AI
  • Whether the robots work safely and reliably outside test labs

Glossary

Nvidia
A US technology company that makes computer chips and AI tools.
Isaac
Nvidia's software platform for building and training AI for robots.

Same story, different eyes

Ad

Add 150 words. Learn 10 a day.

Dicter word books send only the words you learn to your review notes.

More in this topic

RoboticsZDNet Korea

Work That Took 21 Days Now Takes 8: HD Hyundai Robotics and Palantir Advance Physical AI

HD Hyundai Robotics said it is using AI agents with Palantir to shorten on-site problem solving and improve robot programming. After adopting AI agents in August, the average time for related tasks fell from 21.5 days to 7.9 days, about a 63% reduction. The companies expect robot programming costs to drop by more than 50%, and plan to train engineers as on-site solution experts.

Robotics전자신문

Robot Market Expands into Defense: Frequency, Field Testing, and Standardization Challenges Ahead

As the robotics industry expands from manufacturing and logistics into defense, new markets are opening for domestic companies. Key challenges include standardizing technology, conducting field demonstrations, and enabling small and mid-sized firms to enter the defense supply chain. The Army plans to create a robotics and physical AI division in December, and the industry is pursuing communication frequency allocation and common interface standards for operating multiple robots.

Robotics전자신문

Seoul Partners with Seoul National University to Build 'Physical AX Center' to Launch AI Leading City Push

The Seoul Metropolitan Government and Seoul National University signed an agreement on the 30th of last month to build a Physical AX Center at the Seoul AI Hub in Seoul's Seocho-gu, linking R&D, technology demonstration, commercialization, and startups. The first joint project is a "Physical AI Daily Life Living Lab," where assistive, wearable, and care robots will be tested for performance, safety, and user acceptance. The initiative is centered on SNU's H-PAIR national research institute, which receives up to 10 billion won per year in national funding for up to 10 years.

RoboticsZDNet Korea

LG Electronics to Produce Humanoids In-House to Strengthen Competitiveness Through Component Localization

LG Electronics is reportedly producing its semi-humanoid robot CLOiD directly at its Data Factory in Seoul's Yangjae district, after previously relying on contract makers such as AgiBot and Robostar. The company is also moving to bring core components like actuators and computing chips in-house to build its own production capability. Sources say the CLOiD units are currently hand-assembled, with several hundred expected at the Data Factory by year-end.