robothub24
Technology progress

How skilled and how intelligent robots are today

Two capabilities determine everyday utility: hand dexterity and physical AI. In terms of mechanics, robots are approaching human capabilities; the bottleneck is reliability.

Hand: Degrees of freedom comparison Human = benchmark
Human
27
Clone (Muscle)Mfr.
27
Shadow Handoccupied
24
Optimus Gen 3Target
22
Sanctuary AIsec.
21
Figure 03sec.
20
Allegro Hand
16
Optimus Gen 2
11
Full body: degrees of freedom of leading humanoids
Atlas (electric)sec.
56 DoF
Apptronik Apollosec.
44 DoF
Unitree G1max
43 DoF
Optimus
28 DoF
Unitree H1
19 DoF

Manufacturers count differently. Only Shadow is verified from primary sources; the rest are marked as targets or secondary.

Where do we stand? Progress scale from rigid to humanlike
Level 1
Rigid / Industrial

Fixed gripper, pick-and-place

Level 2
Mobile, but rigid

Multi-finger gripper, planned grips

Today
Level 3
Dexterous under supervision

Many joints, in-hand manipulation in demos, dynamic walking

Level 4
Robust Dexterity

Human-centric and reliable in everyday life, walking, standing up

Level 5
Human-like

Fine motor control and whole-body dynamics like humans

Optimists: around 2030

Sergey Levine (Physical Intelligence) sees fully autonomous household robots at the median around 2030, but warns that robust everyday robustness could take another up to ten years.

Skeptics: around 15 years

Rodney Brooks (MIT, iRobot) considers human-like finger dexterity unsolved for approximately 15 years. Bank of America calls the hand the largest cost block, almost 20 percent of a robot.

Source: Rodney Brooks, Why Todays Humanoids Wont Learn Dexterity

Physical AI: the brain

Robots no longer learn individually for each task; instead, they learn through large pre-trained models, similar to ChatGPT for language. The bottleneck is training data, because movement is not available on the internet.

Milestones of recent years
2022
RT-1 (Google)

A robot learns over 700 tasks

2023
RT-2 + Open X

Web knowledge becomes action, 1 million movement data from 22 robots

2024
pi-0

First generalist robot foundation model

2025
GR00T, Helix, Gemini

Foundation models for humanoids, generalization to new environments

2026
Scaling

Race for training data and reliability

Where do we stand? From fixed programming to the ChatGPT moment
Level 1
Hardcoded

Every movement hand-coded

Level 2
Task learned

One task per training

Level 3
Multitask

One robot, many tasks (RT-1)

Level 4
Web knowledge + many robots

VLA models, cross-embodiment (RT-2)

Today
Level 5
Foundation model

Generalization to new environments (pi-0.5, Gemini)

Level 6
ChatGPT moment

Generalist, reliable, economical

Forecast: When will the first fully autonomous robots broadly enter the market?
2026202820302032from 2035
2026/27
Market launch for businesses
Tesla, Figure and 1X deliver first units to companies
by end of 2027
Optimus in sales
Elon Musk, Tesla (target; significantly missed in 2025)
from 2028
Mass adoption industry
Bank of America: breakthrough in manufacturing and logistics
around 2030
Household autonomous (median)
Sergey Levine, Physical Intelligence
2030/32
approximately 1 million per year
Bernstein: approximately 5 percent penetration
from ~2035
Household broad
Bank of America and Morgan Stanley: roughly another decade
~15 years
Counterposition
Rodney Brooks: human dexterity remains unsolved for a long time
Consensus: First commercial scale in factories and warehouses from around 2028, accelerating in the early 2030s. Fully autonomous household robots reaching broad market adoption rather around 2035. The most aggressive timelines come from individuals with direct business interests.
And how do the Chinese see it? Two speeds
China: faster with the body

Mass production is already underway. State goal: Over 10,000 humanoids in commercial deployment by end of 2026, working rather than demonstrating. Unitree: 20,000 units delivered targeted for 2026, following roughly 5,500 in 2025. Around 85 percent of all humanoids delivered in 2025 came from China.

West: ahead on the brain

The best AI models and systems integration come from the USA and Europe. In real-world deployment, Western companies act more cautiously, deploying dozens rather than thousands of robots.

"The bottleneck is embodied AI, not hardware. We are one to three years away from the ChatGPT moment of robotics."

Wang Xingxing, founder of Unitree, August 2025

The point: China's lead is a manufacturing story, not an autonomy story. Even Chinese founders say full autonomy hasn't been solved. On the question of when robots will be fully autonomous, China and the West are close together.

Source: Wang Xingxing (Unitree), WAIC speech August 2025

Market data status: November 2026. Next update: December 2026. Prices come live, other figures are researched anew on a quarterly basis. Sources listed directly with each metric. No investment advice.