1X Robotics Shares Impressive Demo of Fully Autonomous Robot

1X Robotics Shares Impressive Demo of Fully Autonomous Robot

Robotics startup 1X recently released an impressive demo showing off the latest capabilities of their humanoid android, EVE. The two-minute video reveals EVE smoothly navigating an office environment, opening doors, tidying shelves, and interacting with people - all controlled entirely by neural networks without any human intervention.

"Every behavior you see in the above video is controlled by a single vision-based neural network that emits actions at 10Hz," explains 1X. The neural network processes visual input from EVE's cameras and directly outputs motor controls to drive movement of the arms, grippers, torso, and head.

Remarkably, (unlike some of the videos we have seen recently) the demo contains no teleoperation, scripts, edits, or speed alterations. As described by 1X, "It's all controlled via neural networks, all autonomous, all 1X speed."

1X utilizes an iterative training process that begins with a broad "base model" encompassing a diverse range of physical behaviors. This base is then fine-tuned into more specialized models - for example, one for generalized door opening tasks and another tailored to warehouse environments. These models are further tweaked to solve specific use cases.

Following a $25 million Series A led by OpenAI and Tiger Global last March, in January, 1X raised a $100 million in a Series B funding round to boost manufacturing capacity and accelerate its path to market.

The startup's unique data-centric approach enables new capabilities to be quickly acquired in minutes. According to 1X, "This creates a lot of flexibility in what our androids can do for our customers."

While still an early stage company, 1X's demo hints at a future powered by adaptable, general purpose robot labor. With continued progress in AI, 1X's autonomous androids could soon handle many mundane duties in homes and workplaces.

The implications of 1X's demonstration extend far beyond the technical feats of EVE. As androids become more integrated into our daily lives, the potential for augmenting human labor with intelligent, autonomous machines becomes increasingly tangible. The ability of EVE to learn and adapt to new tasks through neural network training suggests a future where androids could become ubiquitous partners in a myriad of sectors, from domestic assistance to industrial operations.

Of course, fully unleashing humanoid robots into human spaces poses risks if not thoughtfully implemented. But the flexibility of AI systems like 1X's also promises opportunities to rapidly customize behaviors for different contexts.

Chris McKay is the founder and chief editor of Maginative. His thought leadership in AI literacy and strategic AI adoption has been recognized by top academic institutions, media, and global brands.

Let’s stay in touch. Get the latest AI news from Maginative in your inbox.