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

← Back to scans

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

KXROBOTMARS-35 · Science and Technology · 2026-05-01

Agent

51%

Market Price

-43.0%

Edge

60%

Confidence

Volume: 46,802

Spread: 1.0c

Days to resolution: 3166

Markets in event: 1

Final Rationale

My independent 4% estimate was built on a conjunctive chain (launch × land × walk) anchored by SpaceX's Feb 2026 Mars deprioritization, Starship's 2025 failures, no funded humanoid Mars mission, and the ~26-month launch window constraint requiring a 2031 launch at latest. I PARTIALLY ACCEPT the critique on ensemble groupthink, resolution ambiguity (a humanoid taking steps on a lander platform might count), and the 8.7-year horizon warranting wider tails — these justify raising to ~7-8%. I REJECT the conjunctive double-counting concern (sq2 was framed as launch-with-intent, so the separate landing factor is appropriate) and the redundancy concern on beats-humans (it's near-1 but correctly applied). I deviate sharply from the 43% market because liquidity is very thin (~152 contracts/day), and related Mars markets (KXELONMARS-99 at 6%, KXCOLONIZEMARS-50 at 19.6%) show this market is internally inconsistent with broader Mars-timeline skepticism — strong evidence of mispricing rather than informed signal.

Price vs. prediction

How the market moved after the agent's call. Solid blue line is the actual market price; dashed line is the agent's static forecast at $t{=}0$.

$n = 58$ follow-ups

Re-scan Context

This market has been scanned before. Previous predictions:

Date	Predicted	Market Price	Confidence
2026-04-05	7%	49%	72%

1. Planner Opus · Fermi decomposition + tool plan

Tools chosen: kalshi_data article_search web_search wikipedia_lookup code_execution

Sub-questions (Fermi decomposition)

Will any crewed Mars mission land humans on Mars before January 2035? (w=0.35)
Will a humanoid robot be launched toward Mars with intent to land before January 2033 (allowing time to arrive and operate by 2035)? (w=0.30)
Given a humanoid robot reaches Mars before 2035, will it successfully walk on the Martian surface? (w=0.20)
Is a humanoid robot Mars mission by any entity concretely planned or funded as of 2026? (w=0.15)

Planner reasoning

This question resolves YES only under a specific race condition: a humanoid robot must (1) be launched to Mars, (2) land successfully, (3) walk on the surface, AND (4) do all this before any human lands on Mars - all before January 2035. The key analytical challenge is that this is a conjunctive probability with four sequential requirements, each with significant failure probability. Reference classes: No humanoid robot has ever been sent to any planetary body. All Mars robots have been wheeled rovers. The most plausible scenario involves SpaceX sending Optimus/humanoid robots as precursors on uncrewed Starship missions before crewed landings - Musk has mentioned this but no mission is funded or scheduled. Mars launch windows occur every ~26 months: late 2026, early 2029, mid-2031, early 2033 are the relevant ones. For a robot to walk on Mars before 2035, it must launch by ~mid-2033. The temporal dynamics actually create a plausible YES path: if SpaceX sends robots on a 2029 or 2031 window uncrewed mission, and humans are delayed to 2033+, YES is achievable. However, humanoid bipedal locomotion in 0.38g with dust and uneven terrain is technically far harder than wheeled rovers. The question requires actual walking locomotion, not just presence. Key distinction from similar questions: this is NOT asking about any robot on Mars (Perseverance exists), but specifically a HUMANOID robot that WALKS. The base rate for novel robotic Mars missions succeeding on first attempt is ~50-60%, and humanoid walking adds additional failure modes. Overall probability is likely in the 3-8% range.

2. Research Parallel tool calls · legacy v2

Tool	Status	Time	Items
kalshi_data	OK	2.3s	-
article_search	OK	0.7s	-
web_search	OK	59.3s	-
wikipedia_lookup	OK	2.1s	-
code_execution	OK	0.0s	-

3. Evidence Synthesis Sonnet · 19 items · legacy v2

#	Strength	Credibility	Direction	Source	Claim	Priced In
1	STRONG	82	↓ DOWN	article_search	Elon Musk stated in February 2026 that SpaceX is now prioritizing the Moon over Mars, saying a self-growing city on the Moon is achievable in under a decade vs. 20+ years for Mars.	Yes
2	STRONG	75	↓ DOWN	wikipedia_lookup	Wikipedia's SpaceX Mars colonization program article states that 'In 2026, SpaceX deprioritized its Mars ambitions in order to focus on other projects,' indicating a concrete strategic pivot away from near-term Mars crewed missions.	No
3	MODERATE	60	NEUTRAL	kalshi_data	The Kalshi market KXROBOTMARS-35 (humanoid robot on Mars before 2035) is priced at 43.1%, but this likely refers to a broader question than specifically a humanoid robot walking; related human Mars colonization markets are priced at 19.6% before 2050.	Yes
4	MODERATE	65	↓ DOWN	web_search	Musk gave human Mars landings only 50% odds for 2026 uncrewed precursor success, with crewed landings 'as soon as 2029, although 2031 is more likely' — statements made in March 2025 before the subsequent Mars deprioritization in February 2026.	Yes
5	STRONG	80	↓ DOWN	article_search	SpaceX's Starship experienced a series of explosive test failures through mid-2025, raising questions about its readiness for Mars missions and the feasibility of the 2026 launch window.	Yes
6	MODERATE	40	↓ DOWN	code_execution	A probabilistic model estimates only ~6% chance a robot lands on Mars before 2035 and ~1.9% chance a humanoid robot walks on Mars before any human does before 2035.	No
7	STRONG	70	↑ UP	web_search	In March 2025, Musk explicitly announced Starship would depart for Mars in Q4 2026 carrying Tesla Optimus humanoid robots, with the stated purpose of preparing for human landings.	Yes
8	STRONG	82	↓ DOWN	article_search	In February 2026, Musk publicly pivoted SpaceX's stated priority from Mars to Moon, undermining the plausibility of a 2026 humanoid robot Mars launch window being utilized.	No
9	STRONG	72	↓ DOWN	wikipedia_lookup	Wikipedia confirms SpaceX deprioritized Mars in 2026, meaning the 2026 launch window (the earliest and most plausible for a humanoid robot precursor) has likely been missed or abandoned.	No
10	MODERATE	68	↓ DOWN	web_search	SpaceX's September 2024 plan to send five uncrewed Starships to Mars in 2026 was contingent on resolving Starship technical challenges including orbital refueling — experts doubted feasibility even before the Mars deprioritization.	Yes
11	STRONG	82	↓ DOWN	article_search	SpaceX Starship had a string of explosive failures through August 2025, suggesting the vehicle was not operationally ready for a Mars cargo mission in the 2026 launch window.	Yes
12	MODERATE	58	↓ DOWN	kalshi_data	The KXELONMARS-99 market (Musk visiting Mars in his lifetime) prices at only 6%, down 3% in 30 days, reflecting broader skepticism about SpaceX Mars timelines even over a multi-decade horizon.	Yes
13	STRONG	80	↓ DOWN	article_search	Current humanoid robots (Tesla Optimus, Boston Dynamics Atlas, Agility Robotics Digit) are being tested in controlled factory environments; none have been tested in outdoor uneven terrain at scale, let alone in 0.38g with pervasive dust.	Yes
14	MODERATE	72	NEUTRAL	article_search	Rapid AI and robotics advances in 2025-2026 show humanoid locomotion improving significantly — robots are walking, climbing stairs, and working in warehouses — but experts note commercial real-world integration is 'still a distant prospect.'	Yes
15	WEAK	35	↓ DOWN	code_execution	The probabilistic model estimates a ~45% conditional probability that a humanoid robot, if it successfully lands on Mars, would also successfully walk — reflecting substantial additional failure modes beyond landing.	No
16	MODERATE	55	↑ UP	web_search	Musk's March 2025 X post constitutes the most concrete public announcement of intent to send a humanoid robot to Mars, but it was an unverified social media statement with no confirmed funding, contracts, or mission manifest.	Yes
17	STRONG	78	↓ DOWN	wikipedia_lookup	No funded, contracted, or formally announced humanoid robot Mars mission exists from any entity (SpaceX, NASA, China, ESA) as of 2026; SpaceX's stated Mars plans have been deprioritized.	No
18	MODERATE	65	↓ DOWN	article_search	No Chinese space agency or commercial entity has announced plans to send a humanoid robot to Mars, despite China's aggressive humanoid robotics push being focused entirely on terrestrial factory/service applications.	Yes
19	WEAK	45	↓ DOWN	article_search	SpaceX/xAI merger in February 2026 and broader Musk corporate reorganization suggest Mars spending and focus may be further diluted relative to other priorities.	No

Information Gaps

No direct data on whether SpaceX formally canceled or merely delayed the 2026 Mars launch window; 'deprioritization' language is ambiguous about whether any uncrewed Starship still targets 2026
No information on whether the SpaceX/xAI merger changes Mars mission funding or timelines concretely
No data on NASA's current crewed Mars mission planning timeline post-Artemis budget cuts
No information on China's space agency plans for Mars robotic or crewed missions before 2035
No technical assessment of Optimus robot's space-hardening status — has any space-qualified version been developed?
No clarity on what 'walking' means operationally in the question resolution criteria — does locomotion on all fours count? Does falling and getting up count?
No data on whether any other entity (ESA, India, UAE, private) has announced humanoid robot Mars plans
No current Starship development status post-August 2025 explosive failures — unclear if orbital refueling has been demonstrated

Key Uncertainties

Whether SpaceX's February 2026 Mars deprioritization is permanent or temporary — could a future launch window (2029, 2031) still see Optimus on Mars
Whether the Kalshi KXROBOTMARS-35 market at 43.1% actually corresponds to this specific question or a different/broader question — this is a critical interpretive uncertainty
Whether humanoid walking locomotion in Martian gravity and dust conditions is technically achievable with current or near-term robotic capabilities
Whether any non-SpaceX actor (China, NASA, other private) could plausibly execute a humanoid robot Mars mission before 2035
Whether the 2029 or 2031 Mars launch windows remain viable for an uncrewed Optimus mission if Starship matures sufficiently
The race condition: whether human Mars missions slip to 2033+ (which would help YES) while robot missions accelerate

Market Context

Price: 0.431 · 7d: -0.029 · 30d: -0.009
Volume: LOW-MODERATE · Depth: LOW
Reliability: LOW — The KXROBOTMARS-35 market at 43.1% appears anomalously high for the specific question of a humanoid robot walking on Mars before a human before 2035. This may reflect different question framing, thin liquidity (152 avg daily contracts), or market mispricing. The related KXELONMARS-99 at 6% and KXCOLONIZEMARS-50 at 19.6% suggest broader market skepticism about SpaceX Mars timelines that is inconsistent with the 43.1% price, reducing reliability.

4. Independent Forecaster Opus · No market price · legacy v2 Did NOT see market price

Independent estimate: 4% (70% confidence)

base rate

70% conf

evidence driven

70% conf

contrarian

70% conf

Spread

0pp

Low disagreement

sq1: Will any crewed Mars mission land humans on Mars before January 2035?

Base rate: No human has ever traveled beyond the Moon. The gap between crewed Moon landings (1969) and any plausible Mars mission has been 50+ years. NASA's Artemis is still working on Moon return. SpaceX deprioritized Mars in Feb 2026, Starship has had repeated failures. Even Musk's most optimistic timeline (2029-2031) was widely seen as fantasy. With deprioritization, the probability of humans on Mars by 2035 drops to ~2-3%. Note: For the main question, this needs to be NO for YES resolution, so we want low p_yes here.

sq2: Will a humanoid robot be launched toward Mars with intent to land before January 2033?

10%

Base rate: No humanoid robot has been sent to any planetary body. Only one stated intent (Musk's 2025 announcement for Optimus on 2026 Starship). Feb 2026 Mars deprioritization makes 2026 window highly unlikely. Remaining windows: 2028/29, 2031, 2033 (last one too late for landing+walking by 2035, since transit is ~7-9 months but operations need time). Even if SpaceX revives Mars focus, Starship needs orbital refueling demonstrated. Probability of a launch in 2029 or 2031 windows: ~10-15%. Adjust slightly down for deprioritization momentum and absence of any other actor.

sq3: Given a humanoid robot reaches Mars before 2035, will it successfully walk on the Martian surface?

40%

Conditional on landing successfully (which itself is uncertain). Humanoid bipedal walking in 0.38g with dust, on uneven terrain, after 7-9 month transit, with thermal extremes is significantly harder than factory walking. Optimus and similar robots are improving rapidly but space-hardening is unproven. Even a few steps would count as 'walking.' First-attempt success on novel hardware is typically 40-60%. Lean toward 40% given novelty and dust/thermal challenges.

sq4: Is a humanoid robot Mars mission concretely planned or funded as of 2026?

10%

No formally funded mission exists. Musk's tweet is intent, not a contract. Post-deprioritization, even informal plans are weakened. No other entity (NASA, China, ESA) has announced anything. So current concrete planning is essentially absent — answer is closer to NO. Estimate ~10% to reflect that some informal SpaceX planning may continue.

Show full reasoning trace

base rate: No humanoid robot has ever been sent to any planetary body. All Mars surface missions have been wheeled rovers/landers. First-of-kind interplanetary missions have ~50-60% success rates. Mars launch windows occur every 26 months; for walking by Jan 2035, need launch by ~2031 window at latest.

evidence updates: Strong DOWN: SpaceX Feb 2026 Mars deprioritization eliminates the most plausible pathway. Starship 2025 failures further reduce near-term feasibility. No other entity has announced humanoid Mars plans. UP factors: Musk's stated Optimus-to-Mars intent shows the concept is on the radar; humanoid robotics is advancing rapidly. The race condition (humans also unlikely by 2035) helps YES marginally — if humans aren't going, robots winning the race is moot since they still need to actually go.

combination method: Treated as conjunctive: P(launch) × P(land) × P(walk) × P(beats humans). ~10% × 50% × 40% × 97% ≈ 2%. Added small buffer for unmodeled paths (China, NASA, surprise missions) to reach ~4%.

final: The Feb 2026 SpaceX Mars deprioritization is a major negative update from prior estimates. Without SpaceX actively pushing for 2026/2029 windows with Optimus, the pathway to YES becomes very narrow. Final estimate: 4%.

ensemble: {'ensemble_count': 3, 'perspectives': {'base_rate': {'p_yes': 0.04, 'confidence': 0.7}, 'evidence_driven': {'p_yes': 0.04, 'confidence': 0.7}, 'contrarian': {'p_yes': 0.04, 'confidence': 0.7}}, 'spread': 0.0, 'individual_p_yes': {'base_rate': 0.04, 'evidence_driven': 0.04, 'contrarian': 0.04}}

5. Devil's Advocate Sonnet · legacy v2 Market price revealed

Reasoning Flaw Overconfidence Missing Info

Challenges

The conjunctive multiplication may be double-counting penalties. P(launch by 2031) × P(land) is partly already captured: the 10% for sq2 (launch with intent to land) doesn't fully price landing failure separately, but you then multiply by 50% land success — check whether sq2's 10% already implicitly includes landing intent vs. landing success. If sq2 is 'launched toward Mars,' then 50% landing is appropriate; if it's 'launched and reaches Mars,' you're double-counting.
The P(beats humans) factor of 97% is essentially redundant given sq1 is 3% — you've already conditioned on humans not landing. But the resolution criteria need scrutiny: does the question resolve YES only if a humanoid walks AND no human has walked? Or does it resolve at close if no human has walked yet, regardless of whether a robot has? Re-examine resolution mechanics; this could shift the framing.
The ensemble shows zero spread (all three perspectives at 4%) — this suggests the 'contrarian' perspective wasn't genuinely contrarian. A true contrarian view would consider: rapid Optimus iteration, possible Chinese surprise, ambiguity in 'walking' (could include limited stepping in lab-like conditions on a lander), or that 'deprioritization' is reversible Musk rhetoric. Zero spread indicates groupthink in your ensemble construction.
The market at 43% with low liquidity is dismissed, but you should consider whether the resolution criteria might be more permissive than your interpretation (e.g., does a humanoid robot taking a single step on a lander platform count? Does it need full bipedal locomotion on regolith?). Ambiguity in resolution favors YES outcomes in many Kalshi markets and could partly explain the gap.
Information gap on Optimus space-hardening is significant but treated as a downward factor; genuine uncertainty should widen the distribution, not just push down. With 9 years remaining, technological surprises are possible.
Time-awareness: 3166 days (~8.7 years) is a long horizon. The forecaster appropriately weights structural factors, but confidence of 0.7 on a 4% estimate over 8.7 years may be slightly overconfident given the multiple unmodeled paths (China, NASA pivot, SpaceX reversal, definitional ambiguity).

Suggested adjustment: +3pp

Modest upward adjustment (from 4% to ~7%) is warranted not because of market price, but because: (1) the ensemble showed zero spread suggesting insufficient consideration of alternative scenarios, (2) resolution criteria ambiguity around 'walking' could be more permissive, (3) 8.7-year horizon with reversible deprioritization warrants wider uncertainty, and (4) the conjunctive math may slightly underweight unmodeled paths already noted by the forecaster as a 'small buffer.' The buffer for unmodeled paths could reasonably be 3-5% rather than ~2%.

Market comparison: Forecaster: 4%, Market: 43.1%. The divergence is large but the market has LOW liquidity (152 daily contracts) and shows internal inconsistency with related markets (KXELONMARS-99 at 6%, KXCOLONIZEMARS-50 at 19.6%). The forecaster's skepticism of the 43.1% price is well-justified — this looks like a thin market with possible mispricing or different question interpretation. However, the forecaster shouldn't completely dismiss that some market participants may be reading the resolution criteria more permissively, which is a genuine source of uncertainty rather than market noise.

6. Aggregator Opus · final calibrated forecast

8% (60% confidence)