Brett Adcock, the founder of Figure.AI, has announced a breakthrough in robotics that could revolutionize the way robots learn and interact with humans. His company is developing a robot called Figure 01, which is set to learn tasks by observing humans, a technique that has proven effective in the field of AI, particularly with models like ChatGPT. This autonomous humanoid robot has learned to make coffee after watching humans. Brett Adcock is calling it the “ChatGPT Moment for Robotics”. Let’s get into the details!
Table of Contents
About Figure 01
Figure 01 unlocks a new era of robotics. It is an autonomous humanoid robot which can learn and perform tasks by observing humans. Standing 5’6″ tall and weighing 60kg, Figure 01 is the world’s first commercially viable autonomous humanoid robot. Equipped with a payload capacity of 20kg, a runtime of 5 hours, and a speed of 1.2m/s, Figure 01 surpasses the capabilities of traditional single-function robots. Figure 01 is designed with a human form factor, bringing together the dexterity of the human form and cutting-edge AI.
Figure 01 Making Coffee
Yesterday, Brett Adcock, founder of Figure Technologies, announced on Twitter that the “ChatGPT moment for robotics” was coming.
we just had an AI breakthrough in our lab
— Brett Adcock (@adcock_brett) January 7, 2024
robotics is about to have its ChatGPT moment
and that moment is happening tomorrow
Today, he shared on Twitter that Figure 01, their autonomous humanoid robot, has learned to make coffee by observing humans. The fact that Figure 01 has learned to make coffee by watching humans is a “ChatGPT moment for robotics,” according to him.
Figure-01 has learned to make coffee ☕️
— Brett Adcock (@adcock_brett) January 7, 2024
Our AI learned this after watching humans make coffee
This is end-to-end AI: our neural networks are taking video in, trajectories out
Join us to train our robot fleet: https://t.co/egQy3iz3Kypic.twitter.com/Y0ksEoHZsW
By watching humans perform this task, Figure 01 could learn the steps involved in brewing a cup of coffee, including pouring water into the coffee maker, waiting for the coffee to brew, and serving the coffee. This is a significant step forward, as it brings the robot closer to performing mundane tasks which humans commonly perform.
Brett Adcock’s Vision: The ChatGPT Moment for Robotics
Brett Adcock, with his deep understanding of robotics and AI, envisions a future where robots can learn tasks simply by observing humans. He draws inspiration from the success of ChatGPT, a language model that made significant strides in conversational AI. Adcock believes that a similar breakthrough in the field of robotics, a “ChatGPT moment for robotics,” would have a profound impact on the capabilities and applications of AI robots like Figure 01.
This vision aims to enable robots to acquire new skills and adapt to dynamic environments without the need for explicit programming or specialized training. Instead, they would learn by observing and imitating human actions, just as ChatGPT learned by analyzing vast amounts of text data. Adcock’s vision represents a paradigm shift in the design and training of robots. This can open up new possibilities for their integration into various industries and domains.
The Power of Figure 01
1. End-to-End AI: Learning from Humans
One of the remarkable features of Figure 01 is its ability to learn tasks by observing humans. Its AI neural networks capture video inputs and generate trajectories. This enables the robot to learn complex tasks such as making coffee. This end-to-end AI approach is groundbreaking because it allows Figure 01 to train on a wide range of human data. Whether it’s making coffee, folding laundry, or warehouse work, we can train Figure 01 to perform various tasks with incredible precision.
2. Scaling to Every Use Case
The scalability of Figure 01 is another key advantage. Figure 01’s potential goes beyond its initial training. As the fleet of Figure 01 robots expands, Figure collected data from each robot, allowing it to retrain the AI system continuously. This iterative process of learning from the collective experiences of the fleet results in improved performance over time. As Figure 01 gathers more data and insights, it becomes even more capable, making it a smart investment for businesses seeking long-term solutions.
The Public Reaction on Reddit (ChatGPT Moment for Robotics)
The Reddit discussion showcases a range of viewpoints and scepticism surrounding Figure 01’s coffee-making abilities. One Reddit user expresses scepticism about calling the AI robot the “ChatGPT moment of robotics.” They argue that without independent verification and more evidence, labelling it as such is simply marketing hype. This user raises a valid point about the need for concrete proof and validation of the robot’s capabilities before making grand comparisons.
To provide some perspective, another user compares the early videos released by Boston Dynamics to the current situation in Figure 01. Like Boston Dynamics’ early videos, the current demonstration may only showcase the robot’s capabilities under specific conditions.
In response to the “ChatGPT moment,” another user suggests that this achievement may be more accurately described as a “transformer moment.” Figure 01’s ability to perform pre-recorded tasks successfully is not on the same scale as the ChatGPT moment.
Others argue that a ChatGPT moment will only occur when successfully shipping such a product, regardless of the timeline. The demo video lacks information about the robot’s other capabilities and raises the possibility that a person could remotely control it.
The concept of a ChatGPT moment in robotics is the latest topic of debate among the Reddit community.
The Future of Humanoid Robots
By replicating human abilities through general-purpose robots, Figure Technologies hopes to usher in a new era where humans and humanoids collaborate to achieve more than they could alone. Only time will tell if such optimistic visions can be fully realized. But the trajectory of progress suggests humanoid robotics may soon start transforming how work gets done worldwide. If successful, it could reframe our relationship with technology and shape economies for generations to come. The age of versatile robot coworkers may be closer than we think.
| Also Read:
- Mobile ALOHA by Standford Opens Doors for Low-Cost Mobile Robotics
- How Google DeepMind is Advancing Robotic Automation with AutoRT, SARA-RT and RT-Trajectory
- EVE Robot, Paving the Way for Next-Gen NEO Robot
- Top 10 AI-Powered Humanoid Robots of 2023
- Tesla Optimus Gen 2: A Closer Look at the Next Generation of Humanoid Robots
- DeepMind’s New AI Robot Learns Independently Without Human Supervision
- AI-Powered Surgical Robots: Will They Replace Human Surgeons?
- The Future of Robotic Process Automation (RPA)






