Intelligence index
Field note 001 / 010 7 min
FIELD NOTE 001

From Jim Fan · NVIDIA GEAR

Three fuels for
embodied intelligence

Robots do not get an internet of actions. They must borrow common sense from human data, rehearse inside synthetic worlds, and pay for experience in physical reality.

A capable robot foundation model will need all three data sources. Each one supplies what the other two are missing.

Jim Fan describes robotics data as a three-part portfolio: internet-scale data, simulation, and real-robot demonstrations. The useful distinction is not simply synthetic versus real. It is whether the data offers breadth, action labels, or physical truth. Today, no single source offers all three.

01 BREADTH

Web data

Watch the world before acting in it.

Human videos, images, text, and instructions expose models to a huge range of objects, activities, and environments. They teach semantic context and common-sense priors: what a mug is, how a drawer usually opens, or what tends to happen after water spills.

Superpower
Diversity at internet scale
Blind spot
No native motor commands

Seeing a hand move is not the same as knowing the joint positions, forces, or control signals that produced the motion.

02 SCALE

Simulation

Practice faster than the clock.

A simulator provides the missing action-to-consequence loop. Policies can attempt a grasp, fail, recover, and repeat across thousands of parallel worlds. The marginal cost of another episode is low, and experience scales with compute.

Superpower
Abundant labeled interaction
Blind spot
The sim-to-real gap

Graphics, contact physics, friction, latency, sensor noise, and environmental diversity can all differ from reality.

03 TRUTH

Real-robot data

Learn where physics cannot be approximated.

Teleoperation and human demonstrations record what actually happens on a specific body: camera frames, robot state, actions, contact, timing, and failures. This is the strongest grounding data for imitation learning and deployment.

Superpower
Embodiment-specific reality
Blind spot
Human time and hardware cost

Collection is bound to wall-clock time, operators, resets, maintenance, safety procedures, and physical wear.

Each source wins on a different axis.

Data source Breadth Action signal Physical fidelity Marginal cost
Web High Low Indirect Low
Simulation Designed Exact Approx. Low
Real robot Limited Exact Highest High

The advantage is the mixing strategy.

The lesson is not to choose one source. It is to assign each source the job it does best, then create a loop in which real failures improve the simulator and targeted demonstrations fill the remaining gaps.

  1. 01
    PretrainBroad priors from web data
  2. 02
    RehearseAction-rich practice in simulation
  3. 03
    GroundCalibrate with real demonstrations
  4. 04
    Deploy & learnTurn failures into the next curriculum

Manual teaching is expensive. That makes data operations strategic.

A hands-on training approach such as teleoperation can produce high-fidelity, task-relevant demonstrations with no sim-to-real gap. Its constraint is economics: every trajectory consumes operator and robot time.

The strategic question is therefore not only how to collect more demonstrations, but how to make each one compound: reuse across tasks, prioritize informative edge cases, automate quality checks, and combine demonstrations with simulation and pretrained visual-language representations.

Keep this

Web data supplies common sense. Simulation supplies practice. Real robots supply ground truth.