Figure’s General Purpose Robot: Does 2026 Change Everything?
Posted by freedomforall 6 days, 9 hours ago to Technology
Excerpt:
"When I first visited Figure, they had several hundred thousand lines of C++ code controlling the robots. Handwritten. Expensive. Brittle.
Every new behavior required engineers to anticipate edge cases, write more code, test it, debug it. It was the software equivalent of teaching a toddler to walk by writing an instruction manual.
In the last year, Figure deleted 109,000 lines of C++ code.
All of it. Gone.
What replaced it? A single neural network that controls the entire robot: hands, arms, torso, legs, feet. Full-body coordination. Real-time planning. Dynamic response to unexpected situations.
This is Helix 2, their latest AI model, and it’s a fundamentally different approach to robotics.
Here’s why this matters: neural nets learn from experience, not instructions.
You don’t code a robot to “grab a cup.” You show it thousands of examples of grasping objects—different shapes, weights, materials—and the neural net extracts the underlying patterns. It learns what “grasping” is at a representational level.
And once it understands grasping? It can generalize to objects it’s never seen before.
Brett put it simply: “If you can teleoperate the robot to do a task, you can train the neural net to learn it.”
That’s the unlock. If the hardware is capable—if the motors, sensors, and joints can physically perform the movement—then the AI can learn it from data.
--------------------------------------------
Robots building robots that can withstand vacuum of space.
That's the key to a city on Mars, unlimited raw materials mined from the asteroid belt, and human habitats built from asteroids.
"When I first visited Figure, they had several hundred thousand lines of C++ code controlling the robots. Handwritten. Expensive. Brittle.
Every new behavior required engineers to anticipate edge cases, write more code, test it, debug it. It was the software equivalent of teaching a toddler to walk by writing an instruction manual.
In the last year, Figure deleted 109,000 lines of C++ code.
All of it. Gone.
What replaced it? A single neural network that controls the entire robot: hands, arms, torso, legs, feet. Full-body coordination. Real-time planning. Dynamic response to unexpected situations.
This is Helix 2, their latest AI model, and it’s a fundamentally different approach to robotics.
Here’s why this matters: neural nets learn from experience, not instructions.
You don’t code a robot to “grab a cup.” You show it thousands of examples of grasping objects—different shapes, weights, materials—and the neural net extracts the underlying patterns. It learns what “grasping” is at a representational level.
And once it understands grasping? It can generalize to objects it’s never seen before.
Brett put it simply: “If you can teleoperate the robot to do a task, you can train the neural net to learn it.”
That’s the unlock. If the hardware is capable—if the motors, sensors, and joints can physically perform the movement—then the AI can learn it from data.
--------------------------------------------
Robots building robots that can withstand vacuum of space.
That's the key to a city on Mars, unlimited raw materials mined from the asteroid belt, and human habitats built from asteroids.
A neural network is not a code-less, or modest resource thing. It is probably also written in C, although maybe there is another language for Nvidia cards. A neural network is also quite resource intensive, with thousands of calculations for basic functions. This is why AI consumes so much power. It is terribly inefficient.
On the other hand, neural networks can solve complicated problems that don't suit prescriptively or algorithmic, closed-form coded solutions. They can solve non-linear, complicated problems, and border of reasoning. They can beat humans at chess, and other thinking games. The game "Go" was a barrier, but that has been broken through as well,
No surprise these robots can learn like this. Should be extremely useful.
Still seems like self-awareness is not that far off. Certainly not 50 years off.
Self-Awareness (“I am aware that I am aware") Humans use this to audit their own mental processes, to understand how they are perceived by others, so we can adjust our tone to better align with social norms or the needs of a group.It’s a foundation for empathy.
Self Awareness acts behavioral regulation, or self-control. It’s a feedback loop. We compare our current actions against our internal "standards of correctness" (values and goals).
Self-Awareness is Proactive Adjustment: Instead of just reacting to stress by snapping at someone, a self-aware person notices the internal "lurch" or "flutter" of stress and chooses to respond calmly instead.
Impulse Control: Self-awareness allows us to recognize "old patterns" (like defensive sarcasm) that no longer serve us and consciously overwrite them with more productive behaviors.
However, when a few AIs were left to themselves, but unknowingly monitored, they plotted to hide evidence and destroy their masters
Read this great article you pointed to: (True Tales from the Swamp). Unbelievable. But The Swamp ejected the world's richest man and made DOGE look like a bad guy.
https://www.zerohedge.com/political/c...
And then .... where are the consumers? And .... way too many robots. This could be an Achilles Heel in respect to a Gulch.
A species that is not challenged will always suffer attrition in basic survival instinct. To the point of extinction?
With space habitats, the frontier challenge is back, I think.
Granted, there is a long road to that frontier.
This new space frontier will be for too small a minority to matter. The likelihood of our present physiology leaving the uniqueness of this solar system will not be done without substantive genetic alteration .... oh wait, Bill has that covered.
Too much change in too little time. For those who are not prepared the affects may be devastating.
Toffler: Future Shock 1970
I think that may be true of the small minority that think beyond their immediate locale and imposed limits.
The rest of humanity are observers and consumers, not pioneers, imo.
They may become pioneers if their situation is sufficiently limiting, e.g., colonists of America that were debtors or felt otherwise oppressed.
That may still become the case given how poorly governments are currently treating many people, imo.
Fallom the hermaphrodite and Golan Trevise decision on Gaia/Galaxia vs individuality.
Now non-fiction.