FAQ
These are the AI models that give humanoid robots their intelligence. Also known as Vision-Language-Action (VLA) models, they enable robots to understand and respond to human actions and language while pursuing the goals they are tasked with. The missing ingredient is training data: vast quantities of real-world human activity data, analogous to the text data that enabled the Large Language Model (LLM) breakthroughs now reshaping knowledge work.
Trace mining partners gather data from real-world task interactions during their regular daily activities. This data is then verified, labeled, and processed to create high value training data for robotic AI servitors.
The
best way to get involved now is to
become
a mining partner and collect training data
during your normal everyday activities. No
special skills or extra work required. TRACE
wearable sensors collect data automatically
during your regular activities and tasks.
Join our community channels to stay current
and position yourself for early access.
Early adopters earn at up to 64x the standard
rate, so early involvement matters.
By
becoming a mining partner, you can contribute to
the data WELL and own an ongoing share of
commercial licensing revenues from the TRACE
training dataset.
Based on current market projections, each
data-hour of quality training data could result
in an annual dividend from an estimated $23 to
$170*. Early adopters will be compensated at up
to 64x the regular rate in the early stages of
data collection, earning 64x hours of revenue
share.
As an example, a person collecting 100 hours of
training data could earn an annual dividend of
$2300* to $17,000*, depending on the adoption
and success of general purpose robotics. As an
early contributor, earnings could be as much as
32x higher or more, up to 64x for the first
public contributors - potentially resulting in
annual returns of up to $1472 - $10880* per hour
spent collecting training data.
*all figures are speculative and depend upon
successful licensing of the WELL dataset to
commercial users.
A market brief explaining the basis of our
segment projections is available for
review in the market projections.