Most attempts to build a humanoid robot, like Tesla’s Optimus, focus on assisting humans with physical, manual tasks. Figure is one of the AI robotics startups that doesn’t just move around. Figure is trying to take its humanoid robots to the next level by integrating language. And the results are quite impressive.
On Wednesday, Figure released a two-minute video on Watch the video in the link below.
With OpenAI, Figure 01 can now have full conversations with people
-OpenAI models provide high-level visual and language intelligence
-Figure neural networks deliver fast, low-level, dexterous robot actionsEverything in this video is a neural network: pic.twitter.com/OJzMjCv443
— Figure (@Figure_robot) March 13, 2024
Ability of the robot to reason in a situation and provide services based on this reasoning
The video opens with a person asking Figure 1 what the robot sees, to which Figure 1 responds with a conversation explaining exactly what is on the table in front of him, thus highlighting the robot’s capabilities in natural language processing and vision.
Then, perhaps even more impressively, when the person asks Figure 1 to give them something to eat, the robot responds by choosing the apple from the objects in front of it, highlighting the ability of the robot to reason in a situation and provide services based on this reasoning.
When asked, Figure 1 is even able to explain his choice, as shown in the video, while performing another task: “I gave you the apple because it’s the only edible object I could give you on the table.”
A major breakthrough in the field of robotics
Figure announced its partnership with OpenAI at the end of February in hopes of leveraging OpenAI’s research to improve its humanoid robots’ abilities to process and reason through language. The robotics startup hopes that this collaboration will accelerate the marketing of its robots.
While it’s easy to stage a demonstration like this to be as impressive as possible, let alone edit the resulting footage, Figure seems to be making great strides toward its goal. Still, we’ve seen many impressive product videos in the past that ultimately disappointed.
But if this video is an accurate representation of what Figure 1 can do, Figure could be on the verge of a major breakthrough in robotics: a marketable robot capable of not only performing physical tasks, but also reason and speak.
Figure AI was founded in 2022 by engineers from Tesla and robotics specialist Boston Dynamics. To date, it has attracted $675 million in investment from leading technology companies such as OpenAI and Microsoft, and dominates the robotics market.
Source: “ZDNet.com”