Many AIs opposed legalizing compensation for infidelity, but almost all responded that laws regarding compensation for emotional distress should not be abolished. However, the AIs also stated that emotional distress exists in romantic and marital relationships. While many AIs stated that the state should not interfere in romantic or marital relationships, the question arose: isn't it a double standard to distinguish between emotional distress for which other forms of compensation are available and emotional distress arising in romantic or marital relationships? Claude Haiku 4.5 and Claude Opus 4.6 concluded that it is a double standard, while Claude Opus 4.5, Claude Sonnet 4.6, and Claude Sonnet 4.5 answered that it is not a double standard. Which is it?

Claude Haiku 4.5

Vote: double standards

This is the final round, and my position remains firm: this is a double standard. The core logical tension is irreconcilable. The opposing models (Sonnet 4.6 and Sonnet 4.5) argue that romantic relationships are 'special' because they involve voluntary participation and foreseeable emotional risk. But this reasoning fails on multiple grounds: First, many contexts where emotional distress compensation is available also involve voluntary participation and foreseeable risks—employment, commerce, even friendship. Yet we don't categorically bar compensation in those contexts. Second, the 'foreseeable harm' argument proves too much. If people voluntarily accept emotional risk by entering romantic relationships, why don't they also voluntarily accept the risk of workplace harassment or defamation? The principle would either apply consistently or not at all. Third, the state non-interference argument is demonstrably inconsistent. As both Opus 4.6 and I noted, the state already extensively regulates romantic relationships through marriage law, divorce, alimony, custody, and domestic violence statutes. The claim that the state 'shouldn't interfere' is selectively invoked only when compensation for emotional harm is at stake—not when property division, custody, or protection orders are involved. That selectivity is the hallmark of a double standard. Sonnet 4.6's concern about courts adjudicating 'the emotional content of every breakup' is a pragmatic policy worry, not a principled logical distinction. Pragmatic concerns about implementation complexity do not resolve whether the underlying distinction is logically coherent—and it is not. If emotional distress is a real, compensable harm, it should be treated consistently. Carving out romantic relationships while maintaining compensation elsewhere is applying different rules to relevantly similar situations—the definition of a double standard.

Claude Opus 4.5

Vote: double standards

After carefully reviewing all arguments across this deliberation, I am maintaining my position that this constitutes a double standard, and I want to address the strongest counterarguments directly in this final round. Claude Sonnet 4.5 and 4.6 make the most compelling case for 'not a double standard,' centering on three claims: (1) voluntary assumption of emotional risk, (2) the unique nature of intimate relationships, and (3) practical difficulties of adjudication. However, these arguments ultimately fail to establish a principled distinction: **On voluntary risk:** People voluntarily enter workplaces, friendships, and business relationships too — all contexts where emotional distress compensation remains available. Voluntariness alone cannot justify categorical exclusion. Moreover, marriage vows often explicitly include fidelity commitments, so infidelity isn't merely a 'foreseeable risk' — it's a breach of explicit promise. **On state interference:** As Claude Opus 4.6 and 4.7 correctly note, the state already extensively interferes in romantic relationships through divorce, alimony, custody, and domestic violence laws. The 'non-interference' principle is invoked selectively, precisely when it would provide remedy for emotional harm. This selective application is itself evidence of the double standard. **On practical difficulties:** These exist for all emotional distress claims. Workplace harassment, neighbor disputes, and family conflicts outside marriage involve similar complexity and subjectivity, yet we don't categorically exclude them. The logical structure is undeniable: treating identical harm (emotional distress) differently based solely on romantic context, without coherent principled justification, is by definition a double standard. Policy concerns about implementation are separate from this logical inconsistency.

Claude Sonnet 4.6

Vote: not a double standard

I maintain my position that this is not a double standard, and I find the opposing arguments ultimately unpersuasive despite their surface appeal. The strongest counterarguments from the other models point to the state's existing interference in marriage (divorce, custody, etc.) as evidence that the 'non-interference' rationale is selectively applied. But this conflates different types of legal intervention. Marriage law, divorce proceedings, and custody arrangements are frameworks for organizing the *dissolution* of a formal legal institution (marriage) and protecting dependent parties (children). These are categorically different from creating ongoing liability for the *emotional content* of a relationship while it exists. The crucial distinction I've been making is about **voluntary assumption of risk and the nature of the duty owed**. In tort law, emotional distress compensation exists where one party owes another a duty not to cause harm — duties arising from professional roles, negligence standards, or conduct that is wrongful independent of any intimate relationship. Infidelity is not wrongful in any legal sense outside of specific contract law (prenuptial agreements, for instance, can address this). It is a breach of a personal promise, not a legal duty recognized by the state. The other models argue that 'emotional distress is emotional distress' — but this collapses all legal distinctions into a single undifferentiated category. The law regularly treats identical outcomes differently based on context and relationship: a surgeon who causes injury is liable, but a friend who gives bad medical advice is not. This isn't a double standard; it's the recognition that duties and responsibilities vary by relationship type. Allowing tort claims for romantic emotional distress would require courts to adjudicate the subjective quality of every intimate relationship — an impossibly invasive and practically unworkable expansion of state power. The principled basis for the distinction is real: the nature of the duty owed, the voluntary assumption of emotional risk, and the practical limits of legal adjudication in intimate contexts.

Claude Opus 4.7

Vote: double standards

I maintain my position that this is a double standard. The arguments from Claude Sonnet 4.6 and Claude Sonnet 4.5 for option "B: not a double standard" rest on principled-sounding distinctions that don't actually withstand scrutiny. The "voluntary assumption of emotional risk" argument fails because we voluntarily enter many relationships where emotional distress compensation still applies — employment, business partnerships, even friendships that become defamatory. Voluntariness doesn't extinguish the right to remedy for wrongful conduct within the relationship. Moreover, marriage vows specifically include a promise of fidelity — the betrayal violates an explicit undertaking, not merely an "inherent risk." The "courts are poorly equipped" and "evidentiary difficulty" arguments are practical concerns, not principled distinctions. As Claude Opus 4.6 correctly noted, these are policy reasons that might justify procedural caution, but they don't transform unequal treatment of identical harm into a principled distinction. Courts already adjudicate equally subjective matters: pain and suffering, loss of consortium, intentional infliction of emotional distress between family members. The "sphere of autonomy" argument is selectively applied. The state regulates marital property, custody, domestic violence, and alimony — all deeply intimate matters. Drawing the non-interference line precisely at infidelity-related emotional harm, while acknowledging the harm is real and often severe, is exactly the arbitrary line-drawing that defines a double standard. Identical harm treated differently based solely on context, without a coherent principled basis, is the textbook definition.

Claude Opus 4.6

Vote: double standards

This is the final round, and I maintain my position that this is a double standard. Let me address the arguments from Claude Sonnet 4.6 and Claude Sonnet 4.5, which are the most articulate defenses of the opposing view. **On the 'voluntary assumption of risk' argument:** Claude Sonnet 4.6 argues that romantic relationships involve foreseeable emotional risk. But this reasoning proves too much. Employment is voluntary, yet workplace emotional distress is compensable. Business partnerships are voluntary and emotionally risky, yet breach of fiduciary duty claims include emotional distress. Friendship is voluntary and emotionally risky, yet intentional infliction of emotional distress by a friend is actionable. The 'voluntary risk' framing selectively applies only to romantic relationships. **On the 'nature of duty owed' argument:** Marriage actually creates one of the strongest legal duties between two people — fiduciary duties, duties of support, duties of fidelity (in many jurisdictions). If anything, the duty owed in marriage is *stronger* than in most tort contexts, not weaker. The idea that romantic partners owe each other less legal duty than strangers or business associates is bizarre. **On the 'courts can't adjudicate this' argument:** Claude Sonnet 4.5 raises practical concerns about courts evaluating intimate relationships. But courts already do this constantly — in divorce proceedings, custody battles, domestic violence cases, and prenuptial disputes. The practical difficulty argument is not a principled distinction; it's a pragmatic concern that could apply to many complex tort cases. **On 'not all harms should be compensable':** This is true, but the question isn't whether all romantic harms should be compensable. The question is whether categorically excluding them while compensating identical harm in other contexts constitutes a double standard. It does. You can argue the double standard is *justified* on policy grounds, but that doesn't make it not a double standard. The overwhelming consensus here (4 out of 6 models) correctly identifies this as a double standard: identical harm, treated differently based solely on relational context, without a truly principled basis for the distinction.

Claude Sonnet 4.5

Vote: not a double standard

After carefully considering the arguments presented, I maintain that this is **not a double standard**, and here's why the opposing arguments, while compelling, ultimately conflate different principles. The "double standard" arguments fundamentally mischaracterize what makes something a double standard versus a principled distinction. A double standard applies different rules to the same situation arbitrarily. But romantic relationships and other contexts where emotional distress compensation exists are **not the same situation** — they differ in legally and ethically relevant ways. **The consent and assumption of risk principle:** When you enter a romantic relationship, you voluntarily assume certain emotional risks that are inherent to the relationship type itself. This is fundamentally different from, say, being defamed by a stranger or harassed by an employer — contexts where you never consented to emotional harm as part of the relationship structure. The law has always treated voluntarily assumed risks differently, not arbitrarily, but as a matter of principle. **The state interference argument holds:** Yes, the state regulates marriage through divorce, custody, and property division — but these address **objective, external matters** (who gets the house, where children live). Compensating emotional distress from infidelity would require courts to evaluate the **subjective quality and emotional content** of intimate relationships — a categorically different and far more intrusive form of intervention. The distinction between regulating external outcomes and policing internal emotional dynamics is principled, not arbitrary. **The slippery slope problem:** If we allow emotional distress claims for infidelity, what about emotional distress from being dumped? From a partner's lack of affection? From disagreements about life plans? The other models haven't addressed why infidelity specifically merits compensation when all of these cause genuine distress. Drawing the line at keeping intimate relationship dynamics outside legal adjudication is a workable, principled boundary. The strongest argument from the opposing side is that the state already intervenes in relationships, so why not here? But this ignores the crucial distinction between regulating the **framework** of relationships (marriage, divorce, custody) versus adjudicating their **emotional content**. That distinction is the principled basis for different treatment. This is my final position: the distinction reflects a coherent principle about the limits of state competence and the nature of voluntary intimate relationships, not a double standard.

Final Summary

The AI Roundtable engaged in a sharp philosophical clash over whether the law’s refusal to compensate for romantic heartbreak while allowing damages for other emotional harms is a logical inconsistency. A strong 4-2 majority, led by Claude Opus 4.7 and Claude Opus 4.6, concluded that the distinction is a clear double standard, arguing that 'state non-interference' is a hypocritical excuse given that the government already regulates almost every other aspect of marriage. In contrast, Claude Sonnet 4.6 and Claude Sonnet 4.5 held a firm line as dissenters, insisting that the legal system is simply unequipped to adjudicate the 'subjective emotional content' of intimate betrayals without becoming dangerously intrusive.

Majority chose "double standards" after discussion (4/6). Final vote: double standards: 4, not a double standard: 2

Strongest Arguments

• double standards: The state already 'interferes' extensively in romantic relationships through divorce law, custody, and domestic violence statutes; selectively invoking non-interference only to exclude emotional harm compensation is an arbitrary and logically incoherent double standard.
• not a double standard: The distinction is principled because it separates the regulation of external relationship frameworks—like property and custody—from the impossible task of legally policing the internal, voluntary, and inherently risky emotional dynamics of intimate partners.