Is this the most autonomous, dexterous robot yet? In this episode of AI News, we cover the massive unveiling of Genesis ENO, a game-changing general-purpose robot challenging the traditional humanoid robot design. Powered by the GENE AI brain, ENO features 20 DoF proprietary hands for human-level dexterous manipulation across manufacturing, logistics, and laboratories.
We also break down Sony AI Ace, the revolutionary autonomous physical agent that just defeated elite human athletes in table tennis under official ITTF rules using deep reinforcement learning and advanced 3D perception. Plus, a look at NVIDIA’s groundbreaking Motion Bricks framework generating robotics physical motion at a staggering 15,000 FPS, and Anthropic’s new Claude Code Artifacts update for developers.
This video highlights recent breakthroughs in robotics and AI, focusing on Genesis's general-purpose robot Eno, Sony AI's table tennis agent ACE, and NVIDIA's Motion Bricks framework. It also covers Anthropix's new \"artifacts\" feature for developers, showcasing a trend toward higher autonomy, dexterity, and real-time generative motion in physical and digital agents.
Is this the most autonomous, dexterous robot yet? Today on AI News, we're taking a closer look at
Eno, the first general purpose robot from Genesis, to see what it can do and how it
adapts autonomously across different tasks and environments. So first, Eno is built to move,
adapt, and learn continuously, but rather than resembling a human entirely,
instead, Genesis designed Eno around capability first, prioritizing mobility, dexterity, and
real-world functionality. As a result, Eno is built with a tower of articulated panels
that are stacked on top of its wheelbase to adjust for height and reach in real-time,
as well as to fold down completely for compact storage while not in use.
And what's more, is that Eno is fitted with two of Genesis' proprietary hands,
each featuring 20 active degrees of freedom, exactly matching the form and function of
human hands, allowing Eno to easily interact with human tools, objects, and environments.
This allows Eno to be truly general purpose and operate wherever people can work,
from factories to laboratories to hospitals and even in homes. And when it comes to brains,
Eno wants to combine AI and hardware into one universal platform. In fact, Eno was designed
in tandem with Gene, which is the robotics-native AI brain from Genesis. So Eno's body and brain
are optimized as a single integrated system. Then, Gene powers Eno's hands with human-level
dexterous manipulation to allow the robot to perform complex, long-horizon tasks,
all with millimeter precision. And in previous demos, its Gene-powered hands were also covered
in biomimetic materials that mimic skin and muscle elasticity to replicate natural surface friction
and give a little to avoid crushing delicate items. But whether or not these biomimetic
materials will be included with Eno is still unknown. Furthermore, unlike most robots that
only execute isolated commands, Gene gives Eno the ability to operate as a true physical agent
that can manage objectives from start to finish. For instance, given a high-level goal, Eno can
understand the context, retain memory, reason through changing conditions, and then dynamically
plan and complete a series of multi-step tasks over an extended period of time. This can be seen
in the real world where Eno goes way beyond single actions and instead manages entire
autonomous workflows, from cooking entire meals to cleaning and more. And by continuously
interacting with its surrounding system, prioritizing tasks, and coordinating with
people and machines, Eno is also able to function much more like an active participant
inside of an adapting situation rather than just another tool that's limited to predefined motions.
But there's also Eno's display, which Genesis is planning to offer as an optional cognitive
interface to show its intent, reasoning, and operational state all in real time. So this
allows everyone interacting with Eno to follow its actions and decision-making process as it
presents its logic on screen. And as for its release, Genesis aims to begin production and
targeted customer deployments by the end of 2026, with Eno first becoming available to industrial
customers. These customers will include manufacturing, logistics companies, and laboratories,
then be followed by service industry customers such as hotels and hospitals, with consumer homes
and outdoor applications following last. Furthermore, Eno is just a first step in the broader
product map of Genesis, with additional embodiments currently in development. But tell us down in the
comments how much the Eno should cost and if there should be a leasing or monthly subscription
package as well. But Sony AI is also entering robotics with the unveiling of ACE, the first
autonomous physical agent to defeat elite human athletes under official International Table Tennis
Federation rules. Published in Nature, this system successfully navigated high-speed adversarial
rallies right at the edge of human reaction time. And to track a ball moving at professional
velocities, ACE uses a hybrid perception network consisting of nine active pixel sensor cameras,
which all capture 3D spatial data at 200 hertz with just a 10.2 millisecond latency, paired with
three event-based vision sensor systems that track ball spin at up to 700 hertz. Then, this hardware
suite feeds into a deep model free reinforcement learning policy that's trained using the Soft
Actor Critic algorithm through 3,000 hours of continuous synthetic self-play and simulation.
And in live testing, the system won three out of five matches against elite club athletes
and has since expanded its record by defeating professional league competitors.
But AI isn't just making robots more precise than ever, it's also making them smoother than ever too,
as NVIDIA Research just introduced Motion Bricks, a real-time generative motion framework for robotics
and gaming that replaces rigid animation graphs with a single neural backbone. It debuted at
SIGGRAPH 2026 and the tech generates physical motion at 15,000 frames per second with a latency
of just two milliseconds, modeling 350,000 distinct motion skills simultaneously. And the system is
trained on the BONUS-C dataset, which consists of 350,000 production-grade motion capture clips
from Bone Studios, and rather than requiring per-task fine-tuning, Motion Bricks instead uses
a unified interface called Smart Primitives, where users simply input high-level navigation targets,
object interaction keyframes, and behavioral style prompts, and the network autonomously
generates the complex physical transitions in real-time. And NVIDIA has already integrated
Motion Bricks directly into its Groot whole-body control stack, which powers global humanoid
research. And in generative AI, Anthropix, Mythos 5, and Fable 5 are still disabled, but the company
did just roll out its artifacts for Clawed Code, which now allows developers to instantly turn a
complete with its code context, tool outputs, and chat history into an interactive, shareable webpage.
The system handles updates automatically, as the local project evolves or code changes,
the generated page even refreshes in real-time at the same URL, while quietly maintaining a clean
version history of all of the sessions in the background. And to spin one up, developers simply
prompt the CLI for a visual asset or ask it for an artifact directly, and then Clawed Code can
provide a secure link that can be launched in a browser or the desktop application and shared
across a team. But Anthropix expects teams to use the feature for practical engineering workflows
like pull request walkthroughs, incident timelines, and architecture audits. Meanwhile, access to
Fable 5 and Mythos 5 is still disabled due to potential jailbreaking concerns by the US.
Tell us what you think Anthropix should do down in the comments below while you watch this video
of BMW's Aeon humanoid hot-swapping its own batteries. Anyways, like and subscribe, and
check out this video here for more of the latest in AI and robotics news, and thanks for watching.