Field guide · public-safe

Robotics S5 upgrade map

Robotics evidence is stronger than a demo-only cycle, but the sector is still S3/S4-heavy. S5 requires a chain: paid deployment, repeatability, runtime/intervention, customer ROI, production economics and financial materiality.

Draft updated 2026-06-13. This page is a working surface for review before the final publishing flow.

Current signal vs required upgrade

AI / model / data stack

NVIDIA Isaac GR00T, Physical Intelligence, Skild AI, Figure Helix and Tesla Digital Optimus make robot AI a trackable infrastructure layer. S5 still needs paid model/software contracts, deployed robot counts, uptime, intervention metrics and revenue economics.

S3/S4

Hardware cost / access

Unitree R1 AIR from US$4,900, G1 from US$13.5K and H2 at US$29,900 lower experiment cost, but do not prove shipments, industrial reliability, gross margin, support cost or customer ROI.

S3/S4

Deployment KPI

Figure BMW shows task-count and runtime evidence. The S5 gap is repeat orders, customer-confirmed ROI/payback, uptime, intervention and margin.

S4

Manufacturing / filing evidence

Tesla capacity-intent language, Figure BotQ production metrics and Leaderdrive supplier filings are meaningful, but each still needs actual output, yield, robot-specific economics or customer confirmation before S5.

S4

What upgrades the sector

The upgrade chain is paid deployment, repeatability, runtime/intervention, customer ROI, production economics and financial materiality. A single demo, logo, price point or capacity headline is not enough.

S5 test

Common misconceptions

  • Cheaper robots lower experiment cost; they do not prove reliability, gross margin or customer ROI.
  • Designed capacity is not current production rate, accepted units or economic utilization.
  • A customer KPI is stronger than a demo, but it is not the same as full unit economics.
  • Supplier revenue growth should not be converted into unnamed humanoid customer exposure.

Sources