Boston Dynamics ATLAS 4.0 Robots NEW NVIDIA AI Is A Game Changer ($40000 HUMANOID)

And the integration of this hardware equips the Atlas to handle complex multimodal AI models, with the upgrade enabling the robot to perform new tasks such as real-time object recognition, autonomous navigation through dynamic environments, and advanced manipulation of tools and objects that it may not have seen yet. Plus, Boston Dynamics developers are actively working hand-in-hand with NVIDIA’s Isaac Lab to improve Atlas’ dexterity and locomotion, thus allowing it to tackle even more intricate activities like assembling components, or even navigating unseen spaces with greater precision in the near future. And this all builds directly on top of Atlas’ existing full-body control system in order to better generalize for real-world challenges with the team already seeing early successes in areas that combine physical dexterity with intelligent decision-making.

And the robot’s Jetson AGX Thor works in two ways, the first of which packs compute of 1,000 sparse 8-bit integer Terra operations per second for lightning-fast straightforward AI jobs like instant object recognition, rapid yes-no decision-making, or spotting and grabbing a falling object, while the second part computes a trillion sparse 8-bit floating point operations per second for even deeper thinking like smoothly navigating obstacles or fine-tuning a grip for the specific amount of force required, and together they allow the Atlas to switch speeds with its thinking. And as for the robot’s real-time computational load, it packs 128GB of RAM, and it’s all powered by Atlas’ onboard 3.7kWh lithium-ion battery pack, which now allows for one hour of mixed-mission operations, including walking, standing, and using tools.

And when it comes to a commercial launch of the Atlas robot, Boston Dynamics is currently planning for late-2026 with initial targets looking at industrial and enterprise use cases, with home robots coming much later. And in terms of the robot’s price, it will probably be well in excess of $100,000. But there’s another robot that’s doing something even more impressive. But before we get into that, first, a message from today’s sponsor. While AI is transforming every industry and reshaping the entire world, are you keeping up with the skills needed to stay ahead? For example, in today’s job market, AI expertise is no longer optional, it’s essential.

Whether you’re looking to advance in your current role or future-proof your career, understanding how to harness AI can set you apart. That’s why Growth School is offering an exclusive three-hour hands-on AI training, designed to help professionals like you master over 25 powerful AI tools. And for a limited time, it’s completely free for the first 1,000 AI news viewers. Whether you’re in marketing, sales, HR, design, product or even finance, it’s not just for tech professionals. AI skills can help you streamline workflows, improve decision-making, and enhance productivity. And there’s more. Everyone who registers will receive a $500 bonus worth of additional resources.

So don’t miss this opportunity to become the top 1% of people developing essential AI skills. Click the link below now and join Growth School’s WhatsApp group for exclusive updates. And going back to robot number two. Aptronics Apollo humanoid robot just got an upgrade with NVIDIA’s Jetson AGX Thor as well to help it in its commercial partnership with Mercedes-Benz, bringing Apollo into their facilities to help with logistics and assembly, while the integration of NVIDIA’s new chip is further supercharging its AI capabilities. In fact, this robot’s deployment aims to automate repetitive, physically demanding tasks to free up human workers to do even more skilled roles.

And Apollo is extremely human-like, standing at 5 feet 8 inches or 1.73 meters, and it mimics human height for seamless integration into workspaces with a weight of 160 pounds or 73 kilograms. It’s sturdy and manageable, and Apollo can even lift a 55 pound payload, which is 25 kilograms, making it ideal for hauling car parts across factory floors, and its battery life lasts a whopping four hours per charge, plus it’s swappable, meaning that it can continue running without a pause, moving at 2.5 miles per hour or four kilometers per hour, allowing Apollo to keep up with production demands efficiently.

And as for its price, the company has said that it’s looking to make the robot roughly around what a new car might cost. So this is probably somewhere between $40,000 to $80,000, with the commercial release not yet having a date, but likely looking at later in 2026. And when the Apollo is compared with other humanoid robots, you can see the way that it handles and carries itself is actually quite advanced. It doesn’t have any of the stutters or momentary pauses that other robots may exhibit. With this one, it seems to be very quick, and it also seems to have a great level of dexterity with its hands mimicking those of humans not having three or four grippers, but five individual digits that seem to have a very wide range of freedom.

And the Apollo robot comes with some extra twists thanks to its advanced force control architecture, similar to the safety advantages of other collaborative robots. Aptronic also features a unique force control architecture in order for safer movements around humans. Plus, it features its own perimeter zone, meaning that the robot detects its own outer zone and adjusts itself to create a safe working environment to protect humans from its actions. And finally, it features its own impact zone, meaning that Apollo can immediately stop when objects enter this detected perimeter zone. And this brings us to robot number one, Neo from 1x.

So this new Neo gamma robot blends autonomy and teleoperation, and it’s able to handle tasks including vacuuming solo, as well as higher level goals like where to move things. And it relies on human input partially to do this, as well as 1x is foundation models, which it used to advance towards full autonomy. And it’s partnered with Nvidia to do this again, using the Jetson AGX Thor for onboard computation, but the robot does even more than chores because it also collects data. And it’s a first mover advantage for the robot, according to 1x is engineers, because this allows for diverse real world interactions further down the line to unlock more meaningful machine intelligence, impacting not only robotics, but also people in the process as later on, this robot will likely be doing things like elder care, as well as other home tasks.

And when it comes to a price for the robot, while this is still undisclosed, it’s likely that this will be well into six figures. And if you look at this robot, tell me in the comments whether or not this almost looks like the original Neo robot, but with just different clothes put on. Check out the gamma in comparison to Neo, and tell me in the comments what you think. But that’s not all because unitary robotics is also pushing the envelope of what’s possible. So check out this video and tell me in the comments what you think and which robot is the most advanced.

When it comes to the speed at which humans will be automated at the workplace, this will probably happen towards the end of 2026. As these companies are planning within the next roughly year and a half to put out these robots to market. But make sure to check out what unitary is also doing. And make sure to comment down below which robot you think is best and which you’d spend your money on. But like and subscribe. And if you want to know more about Japanese robots, check out this video here. [tr:trw].