OpenAI's new Figure 01 robot outperforms Tesla's Optimus

Yesterday, on 13 March, Figure AI, a company collaborating with OpenAI, presented a new video demonstrating the unique capabilities of their joint project - the Figure 01 robot. This robot surpasses the capabilities of Optimus from Tesla, which becomes obvious when watching the video.

Here's What We Know

The video shows a human having a natural dialogue with the robot, asking it to identify what it sees. The Figure 01 robot accurately describes what is on the table and also mentions that it sees the person having a dialogue with it. The man then asks the robot to give him something to eat and the robot hands him an apple responding that it is the only edible item on the table.

All of this is worth watching in the full video to get a proper picture of what is going on. Figure 01 demonstrates a level of autonomy and accuracy that is ahead of Tesla's current achievements with their Optimus robot.

Vision and voice in Figure 01 are handled by GPT-4Vision and ChatGPT technologies respectively, while hearing is provided by WhisperJAX development. Figure AI handled the robot's framework and the model that powers it.

The company claims that all behaviour in the video is based on system learning and is not remotely controlled, meaning that there is no one behind the scenes controlling Figure 01.

But without seeing Figure 01 in person and asking questions, it's hard to verify these claims. There is a chance that this is not the first time Figure 01 has performed this procedure. It might have been the hundredth time, which could explain his speed and fluency.

Source: YouTube