Skip to the card

Card 393 of 9982026-09-04 issue

Observed arrival · 2026-09-04

HumanoidUptime keeps score of what robot deployments refuse to quantify

humanoiduptime.com Observed source
Editorial interest 84/100 Selection signal · not a rating of the site

A public evidence tracker audits humanoid-robot deployments for runtime, failures, human intervention, recovery, and the uptime figures that remain undisclosed.

Landing page captured for the 2026-09-04 issue.

Field notes

The tracker organizes 13 qualifying records across deployment, reliability, and evidence categories, including operator-confirmed activity, manufacturer claims, and independent laboratory tests. Its comparison fields cover runtime hours, availability, human interventions, MTBF, MTTR, task output, autonomy mode, and fleet size. A Figure 02 record cites 1,250+ runtime hours and 90k+ parts loaded but still marks the data needed to calculate availability as unpublished; a Unitree G1 test supplies a bounded laboratory endurance result rather than fleet uptime.

Observed signals

Read the marks

Editorial observations of this landing page, not a rating.

○OpenPublic substance visible
✦PrettyNotable craft visible
●ProPolished or operationally mature
◎NicheUnusually specific use

One card from the complete issue

An Octave Below the Spacecraft

377,806 arrived 1,000 judged 998 catalogued Enter the complete issue
humanoiduptime.com

Landing page observed 2026-09-04. The live site may have changed.