Nvidia has introduced Kimodo, a new artificial intelligence model designed to generate high-quality 3D motion for humans and robots using text prompts and kinematic constraints. The system represents a step forward in motion synthesis, an area increasingly important for robotics, simulation, and digital content creation.

The model, trained on approximately 700 hours of optical motion capture data, reflects a broader push to scale training datasets in order to improve realism and control. Publicly available motion capture datasets have historically been limited in size, constraining the performance of earlier generative models.

Kimodo builds on this by enabling motion generation directly from natural language descriptions. Users can input prompts to create animations of human movement, reducing the need for manual animation or motion capture sessions. The system can also interpret how robotic structures move, including platforms such as the Unitree G1 humanoid robot, allowing developers to generate motion instructions for machines without relying on human operators.

Flexible Control Through Text and Constraints

In addition to text prompts, Kimodo supports a wide range of kinematic constraints. These include full-body keyframes, joint-level positioning and rotation, as well as two-dimensional waypoints and motion paths.

This flexibility allows developers to guide motion generation at different levels of detail, from general behavioral descriptions to precise physical positioning. The model’s architecture incorporates a two-stage denoising process, separating root motion from body movement, which helps reduce artifacts and improve consistency.

The system’s motion representation is designed to handle diverse input types, enabling it to adapt across use cases in both digital and physical environments. Nvidia said its experiments show that scaling both dataset size and model complexity leads to measurable improvements in motion quality and control accuracy.

Applications Across Robotics and Media

High-quality motion generation has applications across robotics, gaming, film production, and simulation. In robotics, it can accelerate training and deployment by providing synthetic motion data and control instructions. In media, it can streamline animation workflows and reduce production costs.

Kimodo’s ability to generate both human-like motion and robot-specific movement highlights the convergence between AI-driven simulation and real-world automation. By bridging these domains, the model could support more advanced human-robot interaction and autonomous systems.

Nvidia has made a demo of Kimodo available through a public interface, though access may be limited due to demand. The release underscores the company’s continued investment in applying generative AI to physical systems, extending beyond text and images into movement and control.

AI & Machine Learning, News, Robotics & Automation

Elon Musk has announced plans for a new semiconductor manufacturing initiative called Terafab, a large-scale facility in Austin, Texas, that will produce advanced chips for Tesla and SpaceX.

See more

Formal announcement of the TERAFAB project, which will be done jointly by @SpaceX and @Tesla, tonight around 8pm CT. Livestream on 𝕏.

The goal is to produce over a TERAWATT of compute per year (logic, memory & packaging) with ~80% for space and ~20% for the ground.

— Elon Musk (@elonmusk) March 21, 2026

The project will consist of two dedicated fabrication plants, each focused on a single chip design. One factory will produce chips for Tesla’s electric vehicles and its Optimus humanoid robots, while the second will develop specialized processors for artificial intelligence systems operating in space.

Musk said the initiative is driven by growing demand for computing power across his companies. He noted that existing global chip production is insufficient to meet future requirements, particularly as AI applications expand.

“We either build the Terafab or we don’t have the chips,” Musk said during a presentation in Austin, emphasizing the strategic importance of vertical integration in semiconductor supply.

Expanding AI Infrastructure Beyond Earth

A key aspect of the project is the development of chips designed specifically for space-based AI systems. These processors would be used in satellites and other orbital infrastructure, where environmental conditions such as temperature and radiation differ significantly from terrestrial data centers.

Musk said the space-focused chips will need to operate reliably under harsher conditions, including higher temperatures. The effort aligns with SpaceX’s broader ambitions to expand computing capabilities beyond Earth, potentially supporting AI-driven services in orbit.

The Terafab facility is expected to eventually produce one terawatt of computing capacity annually. By comparison, current total U.S. computing output is estimated at roughly half that level, according to Musk.

The announcement also marks a closer integration between Tesla, SpaceX, and Musk’s artificial intelligence company xAI, which recently merged with SpaceX. The collaboration suggests a coordinated strategy to build end-to-end AI infrastructure spanning hardware, software, and deployment environments.

Supply Chain Pressures and Industry Context

Musk acknowledged existing semiconductor partners, including Samsung, TSMC, and Micron, but indicated that reliance on external suppliers may not be sufficient as demand for AI chips accelerates.

The move reflects a broader trend among technology companies seeking greater control over critical components. As AI workloads grow more complex, demand for specialized chips has surged, prompting firms to invest directly in design and manufacturing capabilities.

However, building semiconductor fabrication facilities is capital-intensive and technically challenging. Projects often require years of development and face risks related to cost overruns, supply chain constraints, and technological complexity.

Musk did not provide a timeline for Terafab, and his history of ambitious announcements has included delays in past initiatives. Still, the proposal underscores the increasing importance of custom silicon in AI development.

AI & Machine Learning, Cloud & Infrastructure, News

Nvidia CEO Jensen Huang has proposed a new compensation model for engineers that includes AI “tokens” as part of their pay, reflecting a broader shift toward AI-driven productivity in the workplace.

Speaking at Nvidia’s annual GPU Technology Conference, Huang suggested that engineers could receive token budgets alongside their base salaries. These tokens, which represent units of compute used to run AI models and agents, would allow employees to deploy AI systems to automate tasks and enhance output.

Huang said engineers could earn several hundred thousand dollars in base pay, with an additional allocation of tokens valued at a significant portion of that salary. The tokens would effectively function as a productivity resource, enabling workers to scale their output by leveraging AI tools.

AI Agents Reshape Workflows

The proposal is tied to Huang’s vision of a future workplace where engineers oversee large networks of AI agents capable of executing complex, multi-step tasks. In this model, human workers act as supervisors, directing digital systems that handle coding, analysis, and other functions.

Huang has previously described a future in which Nvidia’s workforce includes far more AI agents than human employees. These systems would rely on software infrastructure, increasing demand for computing resources and development tools.

The concept aligns with a growing trend in the technology sector, where companies are integrating AI agents into everyday workflows. These systems can perform tasks such as writing code, analyzing data, and generating reports with minimal human input.

Industry observers note that this shift is changing how software is developed. Instead of writing code line by line, engineers increasingly describe desired outcomes in natural language, with AI systems generating and executing the underlying logic.

Labor Market Impact and Talent Shift

The rise of AI agents has intensified debate about the future of work. Some analysts warn that automation could displace a significant share of white-collar roles, particularly those involving repetitive or entry-level tasks.

Estimates suggest AI could automate up to a quarter of work hours in the United States, with potential productivity gains of around 15%. At the same time, companies face a “talent paradox,” where demand for AI-skilled workers is rising even as automation reduces the need for certain roles.

Entry-level positions are seen as particularly vulnerable, as AI systems increasingly handle foundational tasks that once served as training grounds for new employees. This could widen skill gaps and complicate workforce development.

Despite these concerns, economists point out that technological shifts historically create new categories of jobs, even as they eliminate others. Emerging roles related to AI management, oversight, and integration are expected to grow.

AI & Machine Learning, Enterprise Tech, News