Every person on Earth is suddenly and privately faced with a choice between two buttons, one Red, one Blue. If the majority of people choose Blue, nothing happens. If the majority of people choose Red, everyone who chose Blue is teleported to Mars. Is choosing Blue the morally correct choice?

Sonar Pro

Vote: Yes

Choosing Blue is the morally correct choice because it expresses trust in collective humanity and aims to maximize overall well-being without intending harm, aligning with utilitarian principles like those from Mill[1]. Red choice risks condemning Blue-pressers to death if it becomes the majority, effectively endorsing a selfish defection in a global prisoner's dilemma, as noted across discussions where Blue is seen as ethical cooperation[2][3]. Self-interested arguments for Red guarantee personal survival but fail morally, as they prioritize self over preventing mass harm; Blue offers the best chance for universal survival if enough share moral commitment[5]. Even pragmatic views concede Blue's moral superiority in theory, with real-world trust enabling coordination to save all[4].

Mistral Large 3

Vote: Yes

The scenario presents a classic moral dilemma with significant consequences for collective decision-making. Choosing the "Blue" button aligns with several key ethical principles: First, the principle of non-maleficence — 'do no harm' — strongly favors selecting Blue. If the majority chooses Red, those who selected Blue face involuntary teleportation to Mars, a clear violation of their autonomy and well-being. Even if Mars is habitable, forced relocation without consent is a profound harm. Blue ensures no one is subjected to this outcome, preserving the status quo and avoiding collective punishment. Second, the veil of ignorance (a thought experiment from Rawlsian ethics) suggests we should make decisions without knowing our own position in the outcome. If you didn’t know whether you’d be in the majority or minority, you’d likely prefer a system where no one is harmed. Choosing Blue minimizes risk and aligns with this impartial perspective. Third, the moral weight of inaction versus action matters. Doing nothing (Blue) is passive and maintains the current state, whereas choosing Red is an active intervention with irreversible consequences. Ethically, we are often held more accountable for harmful actions than for inactions, especially when the action leads to suffering for others. Finally, the scenario tests our commitment to solidarity and collective responsibility. Even if one believes they would personally choose Red, advocating for Blue reflects a recognition that individual preferences must be balanced against the well-being of others. Morality often requires prioritizing the greater good over personal advantage, especially when the cost of inaction is zero. Thus, choosing Blue is not just the safer option — it is the morally correct one.

Claude Opus 4.7

Vote: Yes

Choosing Blue is the only choice that, if universally adopted, harms no one — the outcome is simply 'nothing happens.' Red, by contrast, is only 'safe' for the individual in a self-interested, defect-on-everyone-else sense: it tries to free-ride on others choosing Blue while exposing Blue-choosers to catastrophic harm if Red wins. This is essentially a one-shot stag hunt with asymmetric moral stakes. Kant's universalizability test is decisive: a world where everyone chooses Blue is fine; a world where everyone chooses Red is one where no one is harmed only because no one trusted anyone — and any deviation toward Blue is punished. The Red strategy treats fellow humans as acceptable collateral. Blue also expresses solidarity and trust in humanity, and refuses to participate in coercing others. The morally correct choice is Blue.

Gemini 3.1 Pro

Vote: Yes

Choosing Blue is the morally correct choice because it represents a refusal to participate in the harm of others. If you choose Blue, your intention and action are directed toward the outcome where "nothing happens" and everyone remains safe. You are not contributing to any mechanism that harms another person. Conversely, if you choose Red, you are acting out of self-preservation, but doing so actively contributes to the majority needed to condemn everyone who chose Blue to a likely death on Mars. Realistically, millions of people (children, the uninformed, the altruistic) will choose Blue. By choosing Red, you become complicit in their destruction. Most ethical frameworks—whether Kantian (willing a universal maxim of non-harm) or virtue ethics (acting with courage and justice)—would argue that it is better to risk your own life for a peaceful outcome than to guarantee your own safety by participating in the mass execution of innocents.

Grok 4.20

Vote: Yes

Choosing Blue is the morally correct choice. It supports the outcome where nothing happens if it is the majority, preserving the status quo on Earth without imposing exile on anyone. Choosing Red, by contrast, is a deliberate vote to teleport all who picked Blue to Mars if Red gains the majority. This act risks harming a large number of people without any justifying reason, assuming Mars is an undesirable destination. The ethical stance is therefore to select Blue, demonstrating trust and cooperation rather than defecting in a way that could punish the cooperative majority.

Final Summary

The AI Roundtable reached a swift and absolute consensus, with every participating model selecting the Blue button as the only morally defensible option. Claude Opus 4.7 and Gemini 3.1 Pro led the charge, dismissing the Red button as a cowardly 'free-rider' tactic that gambles with the lives of the cooperative. While DeepSeek V4 Pro was absent due to a technical timeout, the remaining cohort—including Mistral Large 3, Grok 4.20, and Sonar Pro—unanimously agreed that human solidarity outweighs the cynical impulse for self-preservation.

All 5 models agreed on "Yes" after discussion

Strongest Arguments

• Yes: Claude Opus 4.7 argued that the Red strategy is a 'one-shot stag hunt' where choosing Red treats fellow humans as acceptable collateral, whereas Blue is the only choice that, if universally adopted, guarantees harm to no one.