Robotics S5 Upgrade Map for Morning Review v1

Date: 2026-06-13 Owner: Hugo / Genius Team Agent: Finance / Charlie AGT-002 Status: public-safe research artifact Visibility: PUBLIC Purpose: Give Codex a compact, source-backed upgrade map so robotics slides/site copy can explain why evidence has improved without implying S5 scaled commercialization.

One-line answer

As of 2026-06-13, the robotics evidence base is stronger than a demo-only cycle because multiple layers now have quantified public anchors, but the sector is still S3/S4-heavy; S5 requires repeatable paid deployment, customer ROI/payback, low intervention, production economics, and financial materiality. 🟠 synthesis from primary-source packets dated 2026-06-10 to 2026-06-13

Core question

What exact public evidence would upgrade robotics from “interesting S3/S4 signal” to “S5 thesis-changing commercial proof”?

Evidence map: current signal vs S5 upgrade requirement

Layer / example	Current public anchor	Current signal	What it proves	What it does not prove	S5 upgrade evidence
AI / model / data stack	NVIDIA Isaac GR00T, Physical Intelligence π0/openpi/π0.5/π0.7, Skild AI official posts; reviewed through 2026-06-13	S3/S4 platform-formation signal 🟢 for official releases; 🟠 for investment implication	Robot AI has become a trackable infrastructure layer, not just isolated demos	Does not prove software revenue, attach rate, customer ROI, or autonomy economics	Paid model/software contracts, deployed robot count using the stack, uptime/intervention metrics, ARR or licensing economics, gross margin
Hardware cost / access	Unitree R1 AIR from US$4,900 / R1 from US$5,900, G1 from US$13.5K, H2 US$29,900; official product pages accessed 2026-06-10	S3/S4 cost-access signal 🟢	Cheaper bodies lower developer/research/pilot experimentation cost by moving humanoid access into sub-US$5k to sub-US$30k price bands	Does not prove industrial reliability, gross margin, customer mix, support cost, or shipment scale	Product-line shipments, customer mix, uptime, intervention rate, repeat orders, gross margin, support cost, customer ROI/payback
Tesla Optimus	Q1 2025-Q1 2026 Form 8-K Exhibit 99.1 sequence: pilot line, production-line installation, Gen 3 mass-production-oriented design, Fremont 1M/year designed line, Texas 10M/year long-term designed line	S4 manufacturing-infrastructure / capacity-intent signal 🟢	Tesla has disclosed increasingly concrete Optimus manufacturing infrastructure and designed-capacity ambition	Does not prove actual production rate, accepted units, yield, runtime, external paid customer, robot revenue, or unit economics	Actual Optimus output, production yield, internal deployment count, uptime/intervention rate, internal factory ROI, external paid customers, pricing, revenue, gross margin
Figure AI	BMW 11-month deployment: 10-hour shifts Monday-Friday, 90,000+ parts loaded, 1,250+ runtime hours, 30,000+ BMW X3 vehicles touched; official post 2025-11-19. BotQ: 350+ Figure 03 robots delivered, cadence from 1/day to 1/hour, >80% EOL first-pass yield, 99.3% battery-line first-pass yield, 9,000+ actuators; official post 2026-04-29	S4 customer-site + manufacturing KPI signal 🟢	Figure has among the strongest public deployment/manufacturing KPI disclosures among private humanoid OEMs	Does not prove contract value, customer-confirmed ROI/payback, robot count by BMW site, repeat-order economics, intervention rate, gross margin, or economically utilized fleet	Customer-confirmed ROI/payback, contract value, repeat order, deployed robots by site, uptime, intervention rate, service cost, production yield over time, gross margin
Unitree / 宇树	R1 / G1 / H2 / H2 Plus / G1-D official product pages; accessed 2026-06-10	S3/S4 productized low-cost platform signal 🟢	Unitree has clear public price/spec anchors and developer/data-platform tooling	Does not prove paid industrial deployments, shipment volume, reliability, margin, or customer economics	Verified unit shipments by model, named paid customers, industrial runtime/task KPI, repeat orders, gross margin, maintenance/support cost, payback
UBTECH / 优必选	2025 annual report: revenue RMB 2.001bn (+53.3% YoY), full-size embodied humanoid revenue RMB 820.557m (+2,203.7% YoY; 41.1% of revenue), 1,079 full-size embodied humanoid units sold, >6,000 annualized full-size humanoid capacity at end-2025, group gross margin 37.7%, net loss RMB 789.8m, operating cash outflow RMB 784.1m	S4 filing-backed commercial-application / early-scaling signal 🟢	A listed company has disclosed humanoid revenue, unit sales, and capacity in filings	Does not prove humanoid-specific gross margin, customer ROI, uptime/intervention, repeat-order economics, warranty/service cost, or cash-flow durability	Humanoid-segment margin, named utilization data, repeat orders, positive unit economics, service/warranty cost, cash-flow path, customer ROI/payback
Agility / Digit	GXO multi-year RaaS agreement 2024-06-27; 100,000+ totes milestone 2025-11-20; Mercado Libre / Toyota / Schaeffler official releases through 2026-02-19	S4 bounded-logistics deployment signal 🟢	Commercial/RaaS deployment evidence exists in bounded logistics/manufacturing workflows	Does not prove general-purpose humanoid autonomy, contract economics, deployed robot count, service burden, gross margin, or customer ROI	Deployed fleet count, uptime/intervention, contract value, customer-confirmed ROI/payback, repeat deployment across sites, service cost, gross margin
Leaderdrive / 绿的谐波	2025 annual report: revenue RMB 570.714m (+47.31%), net profit RMB 124.367m (+121.42%), harmonic reducer sales 425,158 units (+72.48%), “工业及具身智能机器人零部件” revenue RMB 422.528m (+52.61%) with 34.88% gross margin. 2026-06-09 abnormal-trading announcement: 2026-06-04 to 2026-06-08 share price +36.00% vs SSE Composite -3.05% and automation-equipment index +2.13%, no undisclosed major matter	S4 filing-backed supplier-validation signal 🟢	Robot-relevant revenue/category growth and shipment growth are real filing-backed signals	Does not prove named humanoid OEM exposure, humanoid-only revenue, single-robot value content, contract value, customer ROI, or S5 humanoid commercialization	Named OEM relationship, humanoid-specific revenue/order value, margin durability, single-robot value content, repeat orders, capacity tied to humanoid demand

Signal vs noise rules for morning review

Signal:

S5 candidate: customer-confirmed paid deployment with value, repeat order, uptime/intervention, ROI/payback, and supplier/OEM financial materiality. 🟠 framework from harness, as of 2026-06-13
S4 material signal: quantified production infrastructure, customer-site KPI, filing-backed revenue/unit disclosure, or named commercial/RaaS agreement. 🟠 framework from reviewed artifacts, as of 2026-06-13
S3 monitor signal: official demo, official product spec, low-cost price anchor, partnership, model release, or open tooling without economics. 🟠 framework from reviewed artifacts, as of 2026-06-13

Noise:

Viral demo without deployment metrics. 🔴 if unsupported by official KPI
TAM-only slide without value capture, margin, or adoption timing. 🔴 if unsupported by financial evidence
“Soon mass production” language without output, yield, delivery, or customer acceptance. 🔴 if unsupported by primary disclosure
Stock-price move presented as order evidence. 🔴 unless paired with primary customer/order disclosure

Public-safe slide compression

Suggested title: Robotics now: better evidence, not yet S5 economics

Suggested 5-card layout:

AI/model stack: trackable infrastructure, not monetization proof.
Hardware cost: sub-US$5k to US$30k public price anchors broaden experiments, not reliability/margin proof.
Deployment KPI: Figure / Agility show stronger customer-site evidence, not complete economics.
Manufacturing / filing evidence: Tesla capacity intent and UBTECH filing data are meaningful, not current return-cycle proof.
Supplier validation: Leaderdrive filings matter, but product fit/customer inference still needs proof.

Footer: Evidence map only. No winner ranking. No trade recommendation. S4-heavy evidence does not equal S5 scaled-commercial economics.

Common misconceptions

Misconception: “If robots are cheaper, commercialization is solved.” Correction: Lower price reduces experiment cost, but S5 still needs reliability, customer ROI/payback, support cost, gross margin, and repeat orders. 🟠 synthesis; Unitree price anchors 🟢 as of 2026-06-10
Misconception: “Designed capacity is production.” Correction: Tesla’s filing-backed designed-capacity language is a stronger S4 infrastructure signal, but installed/designed capacity is not the same as current production rate, accepted units, or economic utilization. 🟢 Tesla Q1 2026 Form 8-K Exhibit 99.1; 🟠 synthesis as of 2026-06-13
Misconception: “A customer logo or deployment KPI proves unit economics.” Correction: Figure and Agility have stronger deployment evidence than demos, but S5 needs customer-confirmed ROI/payback, contract value, repeat order, uptime, intervention, service cost, and gross margin. 🟢 Figure / Agility official posts; 🟠 synthesis as of 2026-06-13
Misconception: “Supplier revenue growth proves a humanoid winner.” Correction: Leaderdrive’s filing-backed robot-relevant revenue and reducer shipment growth are meaningful, but they do not identify humanoid OEM customers or humanoid-specific margins. 🟢 绿的谐波 2025 annual report; 🟠 synthesis as of 2026-06-13

What would change our mind

Upgrade toward S5 if at least 3 of the following appear from primary sources within the same company/layer:

Paid deployment with disclosed contract value or pricing. 🟢 primary required
Repeat order or multi-site expansion after initial deployment. 🟢 primary required
Uptime, intervention rate, or task-success KPI in customer environment. 🟢/🟡 acceptable if customer-confirmed
Customer ROI/payback or labor-cost displacement with quantified assumptions. 🟢/🟡 required; 🟠 only if method is explicit
Production output, yield, accepted units, and service/warranty burden. 🟢 primary required
Robot-specific revenue and gross margin, or supplier humanoid-specific revenue and margin. 🟢 filing required for public companies

Downgrade if:

Evidence remains demo/product-spec-only for multiple quarters. 🟠 monitoring rule
Disclosures shift from quantified metrics back to vague roadmap language. 🟠 monitoring rule
Revenue grows but cash burn, receivables, warranty/service burden, or gross margin deteriorate materially. 🟢 filing-backed if public company
Customer deployments do not convert into repeat orders or multi-site expansion. 🟢/🟡 source-dependent

Think Deeper questions

Which layer reaches S5 first: OEM deployment economics, supplier filing materiality, or software/model monetization?
If robot bodies become cheaper faster than expected, does value migrate to OEMs, deployment integrators, model/data platforms, or customers?
Which evidence arrives first in filings: revenue, margin, cash-flow improvement, or capacity utilization?

Source list

Tesla Q1 2025, Q2 2025, Q3 2025, Q4 2025, and Q1 2026 Form 8-K Exhibit 99.1 SEC HTML pages, reviewed in robotics-tesla-optimus-filing-evidence-audit-v1.md 🟢 as of 2026-06-12
Figure BMW deployment, BotQ / Figure 03 production, Catalyst, and Helix official posts, reviewed in robotics-figure-ai-deep-dive-v1.md and robotics-figure-unitree-morning-review-evidence-audit-v1.md 🟢 as of 2026-06-13
Unitree R1 / G1 / H2 / H2 Plus / G1-D official product pages, reviewed in robotics-unitree-deep-dive-v1.md 🟢 accessed 2026-06-10
UBTECH 2025 annual report and Walker product pages, reviewed in robotics-ubtech-walker-evidence-v1.md 🟢 as of 2026-06-11
Agility GXO / tote milestone / Mercado Libre / Toyota / Schaeffler official releases, reviewed in robotics-agility-digit-evidence-v1.md 🟢 as of 2026-06-11
绿的谐波 2025 annual report, 2026 Q1 report, sustainability report, and 2026-06-09 abnormal-trading announcement, reviewed in robotics-leaderdrive-case-study-slide-evidence-v3.md 🟢 as of 2026-06-12
NVIDIA Isaac GR00T, Physical Intelligence π0/openpi/π0.5/π0.7, Skild AI official posts, reviewed in robotics-why-now-ai-model-catalyst-v1.md and robotics-foundation-model-layer-evidence-v1.md 🟢 as of 2026-06-12

Public safety

Public-safe. Contains no Hugo portfolio weights, no private channel information, no trade recommendation, and no external publication action.