In July 2025, Skild AI—a robotics startup backed by Amazon and SoftBank—unveiled its groundbreaking technology: Skild Brain, a general‑purpose AI foundation model designed to operate on nearly any robot type. From humanoids to carpet‑climbing quadrupeds and table‑top arms, Skild Brain enables machines to think, adapt, and respond like humans across diverse physical environments. This marks a major leap from task‑specific robotics toward truly universal intelligence.
Unlike prior robotics models limited to scripted movement or narrow tasks, Skild Brain learns through both large-scale simulation and real-world robot data, addressing longstanding data shortage challenges in the field.
Key Demonstrated Abilities:
- Climbing stairs and maintaining balance after being nudged
- Picking up objects in cluttered, unpredictable settings
- Performing household-style tasks such as sorting dishes or operating in kitchen-like environments.
These demonstrations highlight the combination of spatial reasoning, adaptability, and safety mechanisms—including built‑in power limits to prevent excessive force around humans.
Skild Brain is unique for its omni-bodied intelligence: a single model that works across robot morphologies and applications. This contrasts with typical AI approaches that train one model per robot type.
Training, Scaling & Multi-Robot Learning
Data Strategy
Skild tackles robotics’ data scarcity by pre‑training its brain using a combination of simulated environments and human action videos sourced from the internet. This creates a vast, diverse training foundation. Then the model is refined using real deployment data from each robot running on the system.
Shared Learning System
Every robot powered by Skild Brain contributes feedback that sharpens the model. This forms a shared‑brain architecture, where adaptability evolves as more robots are deployed.
Infrastructure Partnerships
Development relies on high-performance infrastructure via partnerships with Hewlett Packard Enterprise (HPE) and NVIDIA. Skild leverages AI-as-a-service solutions and GPU clusters for efficient training and inference across large-scale physical intelligence tasks.
Broader Significance
Skild Brain signals a shift from digitally focused AI to physical general-purpose intelligence. It opens doors to deployment in sectors like logistics, manufacturing, construction, security, and hospitality.
Safety and Human Interaction
The model includes safety constraints and is trained to apply low force, enabling safe interaction with humans—something that many robotic systems struggle to ensure.
Team & Investment
Founded in 2023 by pioneers from Tesla, Nvidia, Meta, and Google, Skild AI has raised over $300M in Series A, valuing the company at $1.5B, with further investment to advance its vision of physical AGI.
Skild Brain embodies a future where robots are not task-bound machines but learning agents capable of thinking and adapting in the real world. With its shared, omni‑bodied AI brain, Skild AI blurs the line between physical robotics and general intelligence—offering a powerful bridge between human-like reasoning and real-world action.
