INDEPENDENT / UNOFFICIALIndependent and unofficial. Not affiliated with, endorsed by, sponsored by, or operated by Unitree Robotics.

PHASE SHIFT

How we read the signals.

PHASE SHIFT provisional v1

The initial estimate was prepared with AI assistance at the operator’s request. This is a provisional editorial framework, not an independently validated scientific scale. The editor can revise the assessment; every change preserves its evidence and rationale.

What the endpoint means

Our “ChatGPT moment” means a broadly accessible system that follows high-level instructions across a useful range of physical tasks, including unfamiliar environments, with sufficiently consistent results to be useful outside demonstrations. It is a reference point for judgment, not a claim that robotics will follow the same adoption path as chatbots.

Scoring anchors

An editorial assessment of readiness for broadly useful, instruction-led robotics in unfamiliar environments. Zero means no general-purpose capability; 25 means mostly narrow demonstrations; 50 means varied tasks demonstrated with substantial operational limitations; 75 means repeated useful deployment across unfamiliar settings with limited intervention; 100 means broad, accessible and reliably useful general-purpose robotics. These anchors are judgments, not measured stages of a fixed engineering programme.

Assessment process

Assess seven dimensions from linked evidence, giving greater weight to transfer and reliability than publicity. Manufacturer demonstrations are attributed claims, not independent validation. The weighted component estimate is rounded to the nearest five points. Editors may override it with a documented reason. Reassessments preserve their sources, component judgments and methodology version. AI assists research and drafting; a human editor remains responsible. No automatic updates or forecasts are in operation.

Weights and uncertainty

Generalisation carries 20%; manipulation, language to action, unseen environments, long-horizon autonomy and reliability each carry 15%; cost and deployment carry 5%. The initial weighted estimate is 39, rounded to 40. Neither the component scores nor the weights are measured scientific constants.

Small numerical differences should not be interpreted as meaningful progress. New replicated evidence, broad deployment or reduced intervention could raise the estimate. Failed replication, hidden supervision, fragility or difficult deployment economics could lower it. A marketing announcement alone does not establish reliability.

Reading the visual

The warm-coloured share of the wave surface represents the current editorial percentage in the grid’s coordinates; the grey share is the remainder. Perspective changes the apparent screen area. Motion and pointer response are illustrative and never change the reading. This is not a map of individual capabilities or a prediction.

Preserving the record

The first assessment is retained as a dated, versioned baseline. Subsequent readings are saved with their actual publication time, previous score, sources, component assessments, methodology and reviewer identity. Reviewer identity is private. Changes to wave speed, height or density do not change the assessment. Article generation never updates the score.

Read the current assessment →