Tesla Optimus public evidence matrix — fresh SEC recheck for Codex packaging
Date: 2026-06-20
As-of: 2026-06-20T05:49:48Z
Owner: Hugo / Genius Team
Research owner: Finance / Charlie AGT-002
Status: SYNTHESIS_CANDIDATE
Visibility: PUBLIC
Output target: slide 11 Tesla evidence module / /robotics/tesla-optimus/ / morning review
Source discipline: 🟢 primary / 🟡 secondary / 🟠 estimate or Charlie synthesis / 🔴 unsupported guess
Public-safety flag: PUBLIC if used as written. Excludes Hugo private portfolio data, watchlist sizing, broker/cash-flow/tax context, private channel checks, supplier rumors, paid-report excerpts, and TSLA buy/sell/hold language.
Related files:
knowledge/robotics-tesla-optimus-sec-quote-claim-bank-2026-06-19.mdknowledge/robotics-tesla-optimus-morning-review-public-evidence-card-2026-06-19.mdknowledge/robotics-tesla-optimus-public-evidence-hygiene-2026-06-19.mdknowledge/robotics-tesla-optimus-evidence-denominator-ladder-2026-06-18.md
0. One-line answer
截至 2026-06-20T05:49:48Z，Tesla Optimus 最适合公开表达为：Tesla 已经有连续 SEC filing-backed 的 Optimus production-line / capacity-intent 证据，强于普通 demo；但公开材料仍没有披露 produced units、accepted units、deployed robot count、runtime/intervention、factory ROI/payback、external paid-customer economics、Optimus revenue、gross margin 或 service burden，因此仍是 S4 manufacturing-infrastructure / capacity-intent signal，不是 S5 scaled-commercial-economics proof。🟢 Tesla Form 8-K Exhibit 99.1 pages filed 2025-04-22 through 2026-04-22; 🟠 Charlie stage classification and denominator synthesis.
Codex-safe compression:
Tesla Optimus: the evidence is real, but the denominator is still capacity design — not current production or economics.
1. Fresh primary-source recheck
Method: from /home/ubuntu/personal-ai/workspace/finance, Python urllib.request fetched five Tesla SEC Exhibit 99.1 HTML pages on 2026-06-20T05:49:48Z, stripped HTML, and searched keyword anchors. The method note is 🟠; the SEC pages are 🟢 primary sources.
| Filing anchor | URL status on 2026-06-20 | Keyword hits | Safe evidence role | Source grade |
|---|---|---|---|---|
| Q1 2025 Form 8-K Exhibit 99.1, filed 2025-04-22 | HTTP 200 | Optimus; robotics; production line; production lines; Installed Capacity | Fremont pilot production-line / internal-factory-use intent. | 🟢 |
| Q2 2025 Form 8-K Exhibit 99.1, filed 2025-07-23 | HTTP 200 | robotics; Installed Capacity | Robotics appears in Tesla official strategic reporting language; not Optimus output proof. | 🟢 |
| Q3 2025 Form 8-K Exhibit 99.1, filed 2025-10-22 | HTTP 200 | Optimus; robotics; production line; production lines; Installed Capacity | First-generation Optimus production-line installation language. | 🟢 |
| Q4 2025 Form 8-K Exhibit 99.1, filed 2026-01-28 | HTTP 200 | Optimus; robotics; production line; production lines; mass production; 1 million; Installed Capacity | Gen 3 mass-production-intended design, first-line preparation, supply-chain readiness, 1M/year planned-capacity language. | 🟢 |
| Q1 2026 Form 8-K Exhibit 99.1, filed 2026-04-22 | HTTP 200 | Optimus; robotics; production line; production lines; mass production; 1 million; 10 million; Digital Optimus | Fremont first-generation line designed for up to 1M/year and Texas second-generation line designed for long-term up to 10M/year. | 🟢 |
Verification caveat: keyword hits prove source reachability and anchor presence; they do not prove actual production, accepted units, utilization, customer economics, or financial materiality. 🟠
2. Claim matrix for Codex
| Public claim candidate | Current status as of 2026-06-20 | Safe wording | Unsafe wording | Signal grade |
|---|---|---|---|---|
| Tesla has Optimus evidence beyond demo clips. | Supported by SEC-filed production-line, Gen 3, supply-chain, and designed-capacity language across 2025-04-22 to 2026-04-22 filings. 🟢 | “Optimus has SEC-backed production-line and capacity-intent evidence.” | “Optimus is commercially proven.” | S4 |
| Tesla is preparing Optimus for mass production. | Q4 2025 says Gen 3 is intended for mass production and first production line / supply chain were being prepared; Q1 2026 gives designed-capacity anchors. 🟢 | “Tesla disclosed mass-production-intended design and line / supply-chain preparation.” | “Tesla has already scaled Optimus mass production.” | S4 |
| Tesla has 1M / 10M capacity anchors. | Q1 2026 gives Fremont first-generation line designed for up to 1M/year and Texas second-generation line designed for long-term up to 10M/year. 🟢 | “Designed / long-term designed capacity anchors.” | “Tesla is producing 1M/year” or “has 10M/year current capacity.” | S4 with caveat |
| Tesla has produced-unit proof. | Not disclosed in reviewed filings. 🟢 absence boundary | “Produced-unit denominator remains undisclosed.” | “Tesla output is already X units” unless future primary evidence exists. | Missing |
| Tesla has accepted / yield proof. | Not disclosed in reviewed filings. 🟢 absence boundary | “Acceptance, FPY, EOL yield, rework and scrap remain undisclosed.” | “Built units are usable deployments” without acceptance evidence. | Missing |
| Tesla has internal deployment economics. | Q1 2025 includes internal-factory-use intent, but reviewed filings do not disclose robot count, task, runtime, intervention, productivity or ROI/payback. 🟢/🟠 | “Internal-factory-use intent exists; deployment and economics denominators are missing.” | “Optimus has proven factory ROI.” | S3/S4 intent; S5 missing |
| Tesla has external paid customers. | Not disclosed in reviewed filings. 🟢 absence boundary | “External paid-customer economics remain undisclosed.” | “Optimus has external customer traction” without named order / value / delivery evidence. | Missing |
| Tesla validates supplier winners. | Reviewed Tesla filings do not name Optimus suppliers or order economics. 🟢 absence boundary | “Supplier value capture requires separate named relationship, BOM/order/revenue/margin evidence.” | “Tesla capacity confirms supplier winners.” | 🔴 unsupported if inferred |
3. Denominator ladder
| Denominator | What it would answer | Current Tesla public evidence as of 2026-06-20 | Why it matters | Stage impact |
|---|---|---|---|---|
| Designed capacity | What scale Tesla is designing toward. | Fremont up to 1M/year; Texas long-term up to 10M/year. 🟢 | This is stronger than demo-only evidence and creates a monitoring denominator. | S4 capacity-intent |
| Produced units | How many Optimus units Tesla built by period/site/generation. | Not disclosed. 🟢 absence boundary | Converts capacity intent into production proof. | S4+ if disclosed, not S5 alone |
| Accepted units / yield | How many units passed usable acceptance. | Not disclosed. 🟢 absence boundary | Separates build volume from usable robots. | S4+ / S5 bridge |
| Deployed units | How many robots work in which factory/task/shift. | Intent disclosed; denominator missing. 🟢/🟠 | Converts production into operating evidence. | S4+ / S5 bridge |
| Productive hours | Runtime, uptime, intervention, reset/failure, safety, task success. | Not disclosed. 🟢 absence boundary | Tests autonomy and reliability in real work. | Strong S5 candidate if paired with economics |
| Economic output | Labor-hour substitution, throughput/quality gain, cost saving, ROI/payback, revenue, gross margin, service burden. | Not disclosed. 🟢 absence boundary | Converts robotics evidence into investable economics. | S5 if disclosed with denominator |
4. Signal vs noise
Signal:
- Five Tesla SEC Exhibit 99.1 pages from 2025-04-22 to 2026-04-22 were reachable on 2026-06-20 and hit relevant Optimus / robotics / production-line / capacity keywords. 🟢
- Production-line installation and Gen 3 mass-production-intended design language are stronger than a viral demo. 🟢
- Q1 2026 provides named-site and quantified designed-capacity anchors: Fremont up to 1M/year and Texas long-term up to 10M/year. 🟢
- Tesla’s capacity-table caveat in prior reviewed artifacts separates installed/design capacity from current production rate. 🟢
Noise unless upgraded:
- Demo videos without robot count, task denominator, runtime, intervention, reset/failure, safety, productivity, or cost-saving data. 🟠/🔴
- “1M / 10M” phrasing without designed-capacity qualifiers. 🔴 if stated as actual current production.
- Produced-unit rumors without primary source, accepted-unit evidence, or deployment denominator. 🔴/🟠
- Supplier winner claims inferred from Tesla capacity ambition. 🔴
- Internal factory intent treated as proven ROI/payback. 🔴
- AI / Digital Optimus language treated as manipulation economics proof without deployment denominators. 🟠/🔴
5. Slide/site-ready block
Suggested title:
Tesla Optimus: capacity intent is real; economics proof is still missing.
Three-card structure:
| Card | Copy | Grade |
|---|---|---|
| What is real | SEC updates show Optimus pilot-line / production-line / Gen 3 / supply-chain / designed-capacity evidence across 2025-2026. | 🟢 |
| What is still missing | No produced units, accepted units, deployed robot count, runtime/intervention, ROI/payback, revenue, gross margin, or external paid customer economics in reviewed filings. | 🟢/🟠 |
| What changes thesis | Produced -> accepted -> deployed -> productive -> economic: output by site/generation, yield, task/runtime denominator, intervention rate, labor/throughput/cost impact, and financial materiality. | 🟠 |
Required footer:
Evidence map only. No TSLA trade recommendation. Designed capacity is not current production. Internal deployment intent is not ROI proof. Tesla capacity does not by itself validate any supplier.
6. What would change our mind
Upgrade toward S5 if Tesla or high-quality customer-side evidence discloses at least three of the following together:
- Produced Optimus units by quarter, factory, and generation. 🟢 needed.
- Accepted units and FPY / EOL yield, including rework / scrap context. 🟢 needed.
- Deployed robot count by Tesla factory, task, shift pattern, and operating duration. 🟢/🟡 needed.
- Fleet runtime, uptime, intervention rate, reset/failure distribution, and safety incidents. 🟢/🟡 needed.
- Labor-hour substitution, throughput / quality improvement, cost saving, or ROI/payback months. 🟢/🟡 needed.
- External paid-customer order with robot count, contract value, delivery schedule, and repeat-order evidence. 🟢 needed.
- Optimus revenue, gross margin, capex burden, warranty / service / support cost, or segment financial disclosure. 🟢 needed.
Downgrade / hold at S4 if future filings keep increasing capacity ambition while produced / accepted / deployed / productive / economic denominators remain absent for multiple reporting periods. 🟠 Charlie framework.
7. Common misconceptions
- Misconception: “Capacity ambition is meaningless hype.”
- Correction: SEC-filed line-preparation and designed-capacity language is stronger than a viral demo and deserves S4 treatment; S4 is not S5. 🟢/🟠
- Misconception: “1M / 10M capacity means production is already scaled.”
- Correction: the reviewed filings support designed / long-term designed capacity, not current production. Tesla’s own capacity caveat separates capacity from current production rate. 🟢
- Misconception: “Produced units alone would prove commercialization.”
- Correction: produced units answer build volume only. Commercialization still needs acceptance, deployment, runtime/intervention, ROI/payback, revenue, margin and service burden. 🟠
- Misconception: “Tesla capacity validates every humanoid supplier.”
- Correction: supplier exposure needs separate named relationship, order/revenue, value content, ASP, margin, architecture durability and replacement-risk evidence. Reviewed Tesla filings do not provide that. 🟠/🔴
8. Think Deeper questions
- Would 1,000 accepted robots with 1,000,000 productive runtime hours and low intervention rate matter more than 10,000 produced robots with no yield/task data?
- If Tesla first proves Optimus inside its own factories, should the KPI be robot revenue or productivity value per deployed robot?
- Which denominator should trigger supplier-file updates first: Tesla output, accepted units, teardown/BOM evidence, or supplier filing revenue?
- How many reporting periods can capacity language stay ahead of deployment/economics before it becomes a negative signal?
9. Source list
- Tesla Q1 2025 Update, Form 8-K Exhibit 99.1, filed 2025-04-22: https://www.sec.gov/Archives/edgar/data/1318605/000162828025018851/exhbit991.htm 🟢
- Tesla Q2 2025 Update, Form 8-K Exhibit 99.1, filed 2025-07-23: https://www.sec.gov/Archives/edgar/data/1318605/000162828025035738/exhibit991.htm 🟢
- Tesla Q3 2025 Update, Form 8-K Exhibit 99.1, filed 2025-10-22: https://www.sec.gov/Archives/edgar/data/1318605/000162828025045861/exhibit991.htm 🟢
- Tesla Q4 2025 Update, Form 8-K Exhibit 99.1, filed 2026-01-28: https://www.sec.gov/Archives/edgar/data/1318605/000162828026003837/exhibit991.htm 🟢
- Tesla Q1 2026 Update, Form 8-K Exhibit 99.1, filed 2026-04-22: https://www.sec.gov/Archives/edgar/data/1318605/000162828026026551/exhibit991.htm 🟢
- Local SEC reachability / keyword verification run on 2026-06-20T05:49:48Z from
/home/ubuntu/personal-ai/workspace/finance; all five pages returned HTTP 200. 🟠 method note; underlying pages 🟢. knowledge/robotics-tesla-optimus-sec-quote-claim-bank-2026-06-19.md🟠 Charlie quote / claim-control synthesis; underlying SEC filings 🟢.knowledge/robotics-tesla-optimus-evidence-denominator-ladder-2026-06-18.md🟠 Charlie denominator synthesis; underlying SEC filings 🟢.
10. Public-safe / internal-only boundary
Public-safe:
- The evidence matrix, denominator ladder, allowed/forbidden wording, slide/site block, misconception section, Think Deeper prompts, and source list above if kept with source grades and no trade language.
Internal-only / exclude:
- Hugo private portfolio data, watchlist sizing, purchase prices, trade rationale, broker/cash-flow/tax context, private channel checks, supplier rumors, paid-report excerpts, proprietary tables, and any TSLA buy/sell/hold or supplier-winner inference unsupported by primary / high-quality evidence.