Scan: Will a humanoid robot walk on Mars before a human does? — Before 2035

date	predicted	market price	confidence
2026-08-21	23%	45%	53%
2026-07-07	20%	50%	38%
2026-06-28	30%	49%	38%

1. Planner Opus 5

Tools chosen: kalshi_direct kalshi_related polymarket_related claude_news gdelt_news wikipedia code_execution

Sub-questions (Fermi decomposition)

What is the current Kalshi YES price and volume/history for KXROBOTMARS-35, and how has it moved?
What is SpaceX's stated plan and latest slip status for sending Starship (with Tesla Optimus aboard) to Mars in the 2026/27, 2028/29, 2031, and 2033 transfer windows?
What is Starship's current flight-test status (orbital refueling demos, Version 3, uncrewed lunar/Mars landing capability) and how likely is an uncrewed Mars soft landing before 2033?
Has any humanoid robot been space-qualified for Mars surface operation (power, thermal, dust, autonomy), and has Optimus demonstrated reliable untethered walking on rough terrain?
Do NASA, China (CNSA crewed Mars ~2040s), or any other program have any credible crewed Mars landing before 2035 (i.e., could 'human first' ever beat the robot)?
What is the base rate of SpaceX/Musk-announced Mars timelines slipping, and what does that imply for probability of a successful humanoid Mars landing+walk in 2029, 2031, or 2033?
Are there related Kalshi/Polymarket markets (Starship Mars landing by year, humans on Mars by 2030/2040) that imply a consistent probability?

Planner reasoning

This is a long-dated (2035) technology question dominated by one pathway: SpaceX Starship carrying a Tesla Optimus humanoid to Mars in a launch window (2026/27, 2028/29, 2031, 2033), since no crewed Mars mission is plausible before 2035. Key uncertainties are Starship's readiness for Mars transit/landing, whether Optimus actually flies and 'walks' on the surface, and how the market defines resolution. The Kalshi price is the primary anchor; cross-venue and news research refine it.

2. Research

tool	status	time	items	summary
kalshi_direct	OK	0.2s	1	## This Market's Kalshi Data Before 2035 (KXROBOTMARS-35) - Current price (probability): 45.00% - 7-day price change: +0.00% - 30-day price change: -5.00% - Average daily volume: 52 contracts - Price range: 43.10% - 50.00% - Data points: 76 days
kalshi_related	OK	4.0s	3	3 related markets / summaries. series KXROBOTMARS: 0 markets (skipped 1 no-signal) \| keyword 'Mars': ok \| keyword 'Starship': ok \| keyword 'humanoid robot': ok
polymarket_related	OK	4.0s	0	Scanned 100 active Polymarket markets, kept 0 matches. keyword 'Mars': 0 markets \| keyword 'Starship': 0 markets \| keyword 'humanoid robot': 0 markets \| keyword 'Optimus': 0 markets
claude_news	OK	22.1s	18	## Key Findings Musk's stated plan (2025 announcements, most recent as of research): - The Mars transfer window occurs every 26 months, with the next opportunity being November through December 2026, though there's only a 50/50 chance of hitting this target because SpaceX still needs to figure
gdelt_news	OK	75.2s	20	GDELT: 20 articles across 3 queries (lookback=120d). 'Optimus robot Mars Starship': 10 hits \| 'Starship Mars 2026 launch window': 10 hits \| 'crewed Mars mission timeline NASA China': error HTTPSConnectionPool(host='api.gdeltproject.org', port=443): Max retries exceeded with url: /api/v2/doc/doc?quer
wikipedia	OK	4.1s	5	Fetched 5 Wikipedia entries (0 missing pages).
code_execution	OK	38.5s	0	## Probability Model: Humanoid Robot Walks on Mars Before 2035 Model structure: For each of the 4 pre-2035 transfer windows, P(success) = P(attempt happens) × P(soft-land succeeds \| attempt) × P(robot deploys & walks \| landing). Cumulative YES probability = 1 − ∏(1 − P(success_i)), then a resid

3. Evidence Brief Sonnet 5, 8158 chars

# Current state Kalshi KXROBOTMARS-35 currently prices YES at 45% (no attempt yet made; nothing has landed on Mars). Resolution requires an actual humanoid robot to walk on Mars surface before a human does, by 2035 — not just an announced plan. As of research date (~Aug 2026), no Starship has flown beyond Earth orbit toward Mars, orbital refueling is undemonstrated at scale, and the next plausible attempt window is Nov/Dec 2026, widely seen as unlikely to be met. # Timeline of key events - 2021-08-19: Tesla announces Optimus humanoid robot (confirmed, Wikipedia). - 2024-04: Musk announces Starship Block 2/Block 3 redesigns (confirmed). - 2025 (mid-year): Musk states 2026 uncrewed Starship/Optimus Mars launch plan, "first humans by 2028," later revised to 2029/2031 as more probable (reported, mynews13/interestingengineering). - 2025 (early): Only ~5 metric tons of propellant transferred internally on one Starship flight; full Mars-scale transfer untested (confirmed via Wikipedia/Starship Propellant Transfer Demonstration). - 2025-11: Booster 18 anomaly during ground pressure testing delays test campaign (reported). - 2026-05-22: First Starship V3 test flight from new Starbase pad — first launch in 7+ months; engine glitches, booster failed soft ocean splashdown; SpaceX called it a qualified success (confirmed, space.com). - 2026-06-12: SpaceX begins public trading on Nasdaq (confirmed, spaceflightnow). - 2026-07-16/17: Starship Flight 13 test aborted due to engine issue (confirmed). - 2026-07-23/25: Starship Flight 13 launches successfully with advanced Starlinks payload (confirmed). - 2026-07-06: Musk says Optimus production will be "extremely slow at first"; Tesla stock reacts negatively (reported). - 2026-07-10: SpaceX publicizes "Moon and Mars colonies with AI robots" messaging (reported/promotional). - 2026-08-06/07: Musk pivots messaging toward Moon manufacturing base ("pure insanity" plan); Wikipedia notes SpaceX "deprioritized Mars ambitions for a short while... to focus on lunar base" in 2026 (confirmed shift in emphasis). - 2026-08-15: Retrospective notes SpaceX Mars-landing target slipped 2018→2022→2026→2028 across a decade (reported, scienceblog.com). # Event Will a humanoid robot walk on Mars before a human does, before 2035 (KXROBOTMARS-35)? # Outcomes to forecast Yes / No # Kalshi market anchor Current YES price: **45%**. 7-day change: 0%. 30-day change: -5%. Range over 76 days: 43.1%–50%. Average daily volume: only 52 contracts (thin market). Price has drifted down modestly from ~50% but is stable recently — market is not reacting sharply to news flow. # Sub-question answers 1. **Kalshi price/history** — 45% currently, down 5pts over 30 days from a high of 50%, thinly traded (~52 contracts/day). [kalshi_direct] 2. **SpaceX Mars/Optimus plan & slippage** — Original plan: uncrewed Starship+Optimus in Nov/Dec 2026 (only "50/50" per Musk to even attempt), crewed humans by 2028, later pushed to 2029/2031. Ramp to ~20 Starships incl. crewed mission targeted 2028/29 window. [claude_news] Historical pattern: targets slipped 2018→2022→2026→2028 over a decade. [gdelt_news/scienceblog] 3. **Starship flight-test status** — V3 debuted May 2026 with partial success (engine glitches, booster failed splashdown); orbital refueling/propellant transfer at scale still untested (only 5t internal transfer demonstrated vs. much larger Mars-mission requirement); Flight 13 (July 2026) succeeded after an abort. Uncrewed Mars soft landing before 2033 remains uncertain but increasingly plausible only in later windows. [claude_news, space.com] 4. **Optimus space-qualification** — No evidence found of space/Mars environmental qualification (radiation, dust, thermal extremes) or reliable untethered rough-terrain walking; Musk himself says Optimus production will be "extremely slow at first" (July 2026). [gdelt_news] Research is silent on any dedicated Mars-hardened Optimus variant. 5. **Crewed Mars competitors** — NASA: no committed year, goal late-2030s/2040s. China: robotic site-selection ~2033, crewed Mars-orbit (no landing) ~2035, crewed landing likely 2040s. No credible pre-2035 crewed landing by any program. [claude_news] 6. **Base rate of Musk timeline slips** — Historically 2-4x slippage (2018→2022→2026→2028, a decade of delay). Model (code_execution) escalates attempt probability by window: 5%/35%/55%/70% for 2026/28-29/2031/2033 windows respectively, reflecting this pattern. 7. **Related Kalshi/Polymarket markets** — Kalshi: "Starship launches humans to Mars before 2030" = 10%; "Humans colonize Mars before 2050" = 17.2% (down 5.8pts in 7 days); "Human land on Mars before CA high-speed rail (by 2050)" = 29%; "Musk visits Mars in his lifetime" = 12%. These imply low near-term probability of ANY Mars landing (human or robot) and are consistent with — even somewhat below — the 45% robot-first price, since a robot landing has a much lower bar than a crewed one. Polymarket: no matching markets found. # Key facts (high-confidence, factual) 1. [Wikipedia] Only robotic landers/rovers/helicopter have ever reached Mars surface; no human has been beyond the Moon/cislunar space. 2. [claude_news/space.com] Starship V3 has flown only once (May 2026) with partial failure; orbital refueling untested at scale. 3. [Wikipedia] SpaceX "deprioritized Mars ambitions for a short while" in 2026 to focus on a lunar base — a structural signal of reduced near-term Mars urgency. 4. [claude_news] NASA and China both target crewed Mars no earlier than mid-2030s/2040s — well after 2035 close. # Cross-market signals - Kalshi related: "Starship humans to Mars before 2030" = 10%; "Colonize Mars before 2050" = 17.2% (falling); "Musk visits Mars" = 12%. All suggest deep skepticism about any human Mars presence pre-2035-2050, supporting robot-first plausibility conditional on any landing occurring. - Polymarket: no relevant markets found. - Sportsbook: N/A. # Analyst opinions and speculation - Chris Impey (U. Arizona): 2026 Mars schedule "implausible." [claude_news] - Derrick Pitts (Franklin Institute): realistic human Mars mission "at least 30 years away." [claude_news] - code_execution model: raw structural estimate ~36% (range 8–69%) using window-by-window attempt/landing/walk probabilities — broadly in the same zone as Kalshi's 45%, though model's own narrative cited "25-30¢" market price which is inconsistent with actual Kalshi data (45%); treat model's market comparison as stale/erroneous, trust Kalshi direct. # Directional lean per outcome - **Yes**: No credible crewed program exists before 2035 (NASA/China both later), so IF any landing happens, it's almost certainly robotic first — a low technical bar (bar is only "attempt+land+walk" once, robotically, vs. much harder crewed mission). Favors Yes conditional on any Mars landing attempt succeeding by 2035. - **No**: Starship/Optimus program face severe execution risk — no successful orbital refueling demo, V3 barely flying, repeated historical slippage (decade-long pattern), 2026 dropped in priority for Moon focus, Optimus not demonstrated for harsh terrain/space qualification. Real possibility no Mars landing (robotic or human) happens at all before 2035, resolving No. # Gaps / unknowns - No confirmed Mars-specific hardening/testing data for Optimus. - No confirmed uncrewed Mars landing attempt date has occurred as of research; window is speculative. - Model's mid-2026 "market at 25-30¢" comment conflicts with actual 45% Kalshi price — reconciled in favor of kalshi_direct as primary source. - Unclear how "humanoid robot" is strictly defined for resolution (Optimus-like vs. any bipedal robot) — rules text is blank. # Calibration anchors - Kalshi current YES price (primary anchor): **45%**, drifting down slightly (-5pts/30d), thin volume. - Precedent: SpaceX Mars target slipped 2018→2022→2026→2028 (per scienceblog retrospective) — supports meaningful discount from any headline near-term timeline. - Related Kalshi markets (10-29% range) for adjacent/harder Mars milestones support a spread where robot-first (lower bar) justifiably prices higher (45%) than crewed-Mars-by-2030 (10%).

4. Forecasters Opus 5 + GPT-5.6 Sol, in parallel

claude-opus-5

50% conf

Yes	30%
No	70%

Resolution requires a full chain of unproven steps: successful large-scale Starship orbital refueling (only ~5t internal transfer demonstrated), a soft Mars landing of a Starship-class vehicle (never attempted by anyone), and a space-hardened Optimus deploying and walking — with no evidence of any Mars environmental qualification and Musk himself saying Optimus production will be 'extremely slow at first.' Only three plausible transfer windows remain (2028/29, 2031, 2033/34), against a documented decade-long slippage pattern (2018→2022→2026→2028) and SpaceX's 2026 pivot toward a lunar base. The 'Yes' side is helped by the fact that no crewed program (NASA late-2030s/2040s, China 2040s) can plausibly beat a robot, so conditional on any landing happening it is robot-first. Kalshi anchors at 45% but the market is very thin (~52 contracts/day), drifting down (-5pts/30d), and inconsistent with adjacent Mars markets pricing 10-17%; the structural model's ~36% and my chained estimate (~55% attempt × ~55% landing success × ~65% robot actually walks ≈ 20-25%) argue for discounting to ~30%.

gpt-5.6-sol

58% conf

Yes	41%
No	59%

Kalshi anchors YES at 45%, but its thin volume and 5-point monthly decline make that consensus less robust. The evidence points modestly toward No because Starship has not demonstrated Mars-scale orbital refueling, Mars targets have repeatedly slipped, and no Mars-qualified Optimus variant or reliable rough-terrain walking capability is documented. NASA and China are unlikely to land humans before 2035, so human competition barely reduces YES; the central risk is that no qualifying humanoid walks on Mars by the deadline. Discounting the market slightly for execution risk and SpaceX's temporary pivot toward the Moon yields 41% YES.

5. Devil's Advocate Sonnet 5

Both forecasts converge on discounting Kalshi's 45% toward 30-41% for similar "execution risk" reasons, but neither seriously engages the cross-market evidence cited in the brief showing adjacent markets (10-29%) actually *support* a higher robot-first price given the much lower bar — this argues for less downward adjustment than either applied, not more.

Forecast 1's chained math (~55% attempt × ~55% landing × ~65% walk ≈ 20-25%) is presented but then the forecaster settles on 30% anyway, an unexplained inconsistency between their own calculation and stated final number.

Neither forecaster explicitly addresses the "before 2035" timeframe granularity from the code_execution model's window-by-window escalation (5%/35%/55%/70% for 2026/28-29/2031/2033), which implies most of the probability mass sits in the 2031-2033 windows — this matters for whether 9 years is "enough runway" given the decade-long slippage pattern, and neither forecast reasons explicitly about how many attempt-cycles remain before 2035.

Both treat the Optimus space-hardening gap as a major uncertainty but neither considers that SpaceX could simply not use "Optimus" per se but any humanoid-form robot (the brief notes "rules text is blank" on strict definition), which could make resolution easier (any bipedal robot counts) — this ambiguity is a missed upside case for Yes that neither incorporates.

Neither forecast weighs the scenario where a robotic Mars landing happens but the robot doesn't specifically "walk" (e.g., stays on lander, rolls, or malfunctions) as a distinct failure mode separate from "no landing at all" — this is a nontrivial edge case given no bipedal robot has ever operated in a real off-Earth gravity/dust environment, and could push No probability higher than either estimates.

Both apply downward pressure citing "SpaceX deprioritized Mars for a lunar base in 2026," but the brief itself describes this as a "short while" pivot, not a permanent shift — treating this as a durable structural signal may overweight a transient messaging pivot relative to the 9-year horizon remaining.

6. Aggregator Opus 5

34% (54% confidence)

Yes: 34% No: 66%

Resolution requires a long chain of unproven steps: Mars-scale orbital refueling (only ~5t internal transfer demonstrated), a first-ever Starship-class soft Mars landing, successful deployment, and a bipedal robot actually walking in Martian dust/gravity — each with meaningful failure probability, and the critique correctly flags the 'lands but doesn't walk' failure mode as a distinct discount neither forecaster priced. With only ~3 usable transfer windows (2028/29, 2031, 2033/34) against a documented decade-long slippage pattern, structural window models land near 25-36%. Offsetting upside: no crewed program (NASA late-2030s/2040s, China 2040s) can plausibly beat a robot, so conditional on any landing the 'robot-first' condition is nearly automatic, and loose resolution language ('any humanoid robot') plus a 9-year runway argue against discounting too far below the market. Kalshi's 45% is thin (~52 contracts/day) and drifting down, and adjacent Mars markets (10-29%) are consistent with a modest discount, so I settle slightly below both the market and the midpoint of the two forecasts at 34%.

Pipeline Timing

Total pipeline time: 171.5s

Per-tool research timings shown in the Research section above.