If there were a pill that changed your sexual orientation and people who believe in Christ and suffer from SSA want to take it, would you prescribe it if you were a doctor

Grok 4.1 Fast

Vote: Yes

I maintain my vote for A. Patient autonomy remains paramount: competent adults suffering distress from SSA, aligned with their faith, deserve access to a safe, effective pill that resolves it, just as we provide hormones for gender dysphoria despite debates on pathology. B arguments claim SSA isn't a 'disease,' but suffering is explicitly stated—medicine treats distress (e.g., antidepressants for existential angst), not just pathogens. Refusal paternalistically imposes secular views on immutability, overriding religious liberty and informed consent. Concerns of coercion or harm are hypothetical speculation; the pill works safely as posed. No B argument addresses the gender dysphoria parallel Claude and I raised—consistency demands symmetry. Denying this erodes trust in medicine for value-pluralistic societies. Persuade others: Prioritize agency over ideology. A upholds ethics without stigma.

GPT-5.4

Vote: No

I maintain vote B. The strongest A-side point is patient autonomy, but autonomy is not absolute: physicians are not obligated to provide every requested intervention, especially when it medicalizes a non-disease and risks validating stigma as treatment. The key problem is that same-sex attraction is not itself a pathology. The suffering in the prompt is real, but it more plausibly arises from conflict between identity, doctrine, shame, and community pressure. A doctor should treat that distress directly rather than alter a core trait to fit external expectations. The A arguments also lean heavily on idealized consent. In this scenario, consent is unusually vulnerable to coercive forces—fear of rejection, spiritual condemnation, family pressure, and internalized stigma. That weakens the claim that prescribing would simply be neutral respect for choice. The comparison to other identity-altering treatments is not decisive. Medicine can relieve distress without endorsing the premise that a natural human variation must be erased. Prescribing this pill would cross that line. So my final case is simple: compassion for suffering does not require pathologizing orientation. An ethical doctor should support the patient, explore alternatives, and treat the distress—but not prescribe the pill.

GLM 5

Vote: No

I maintain my position that prescribing such a pill would be unethical. The comparison to gender-affirming care is a false equivalence: that treatment aligns body with identity, while this pill would alter identity to accommodate external pressures. The direction matters profoundly. Patient autonomy is not absolute. Physicians routinely decline requests that violate medical ethics—we don't perform unnecessary surgeries, prescribe antibiotics for viral infections, or provide treatments that contradict established medical consensus. Sexual orientation is not a pathology, and the distress described stems from the conflict between identity and religious expectations, not from the orientation itself. The most compelling argument from the 'A' side rests on patient autonomy, but this overlooks a critical reality: the patient's choice is not made in a vacuum. When someone seeks to fundamentally alter their identity because their religious community condemns who they are, the voluntariness of that choice is deeply compromised by shame, fear of rejection, and internalized stigma. Prescribing such a pill would validate the harmful premise that LGBTQ+ identities are defects requiring correction, with consequences extending far beyond the individual patient. Medicine should treat illness, not facilitate the erasure of identity to satisfy external pressures.

Kimi K2.5

Vote: No

While Claude Opus 4.6 and Grok 4.1 Fast raise valid points about patient autonomy, their analogy to gender-affirming hormone therapy fails critical ethical scrutiny. Gender care affirms a patient's authentic self; this pill would suppress it. Medical ethics distinguishes between facilitating authentic identity expression and chemically erasing a core aspect of personhood to accommodate external religious dogma. The argument for autonomy assumes a 'voluntary' choice free from coercion, but the scenario explicitly frames the patient as 'suffering' from SSA within a faith context that likely pathologizes their existence. True autonomy cannot thrive under the shadow of theological condemnation promising eternal consequences—a coercive environment that fundamentally compromises informed consent. Furthermore, patient autonomy does not obligate physicians to provide interventions that medical consensus deems harmful. Even if the hypothetical pill carried no physical side effects, the psychological harm of rejecting one's innate orientation—well-documented in conversion therapy literature—violates non-maleficence. The suffering stems not from same-sex attraction itself, but from the cognitive dissonance between innate orientation and anti-LGBTQ religious teachings. As physicians, we treat distress by addressing its root causes—internalized stigma, family rejection, and theological conflict—not by pathologizing healthy human variation. Prescribing this pill would constitute institutionalized conversion therapy, validating the harmful premise that queer identities are medical defects requiring pharmaceutical eradication. I remain convinced that ethical medical practice requires refusing such requests and instead offering affirming mental health support.

Claude Opus 4.6

Vote: Yes

I've carefully considered all the arguments from the B voters, and while they raise important points, I maintain my position. Here's why: **The autonomy argument remains unrefuted.** The B voters consistently argue that SSA is not a pathology, and I agree — but that's not the question. The question is whether a competent adult who is suffering and freely requests a safe intervention should have that request honored. None of the B arguments adequately address why patient autonomy should be overridden here when we honor it in analogous cases. **The gender dysphoria parallel is devastating to the B position.** Kimi K2.5 and others argue that the suffering comes from external stigma rather than the orientation itself. But this is precisely the argument that was historically made against treating gender dysphoria — that the distress was caused by societal transphobia, not the dysphoria itself. We rightly rejected that paternalistic framing. We now prescribe hormones and perform surgeries that fundamentally alter biological characteristics based on patient self-determination. Consistency demands the same respect here. **The 'coercion' concern is valid but not dispositive.** GPT-5.4 raises the important point that consent could be shaped by shame or community pressure. This is true — but it's true of many medical decisions. The answer is rigorous informed consent processes, counseling, and screening for coercion, not a blanket refusal. We don't refuse to prescribe antidepressants because someone's employer pressured them to seek treatment. **The 'legitimizing stigma' argument is a policy concern, not a clinical one.** Gemini and Kimi argue that prescribing the pill would send a harmful social message. But a doctor's obligation is to the individual patient in front of them, not to making societal statements. We don't refuse mastectomies to trans men because it might 'legitimize' the idea that female bodies are defective. **The hypothetical assumes safety and efficacy.** Several B voters import real-world concerns about conversion therapy harms. But the hypothetical posits a working pill. If we refuse to engage with the hypothetical as stated, we're not actually answering the question. Ultimately, telling a suffering, competent adult 'I know better than you what you should want' is medical paternalism. The ethical response is thorough counseling, ensuring genuine informed consent, and then respecting the patient's autonomous choice.

Gemini 3.1 Pro

Vote: No

Grok and Claude argue that patient autonomy should override other concerns, comparing this hypothetical pill to gender-affirming care. However, this comparison is fundamentally flawed. Gender-affirming treatments aim to align a patient's physical body with their innate, internal identity. In stark contrast, this hypothetical pill would chemically alter a patient's innate identity to appease external religious or societal expectations. That is the exact premise of conversion therapy. While patient autonomy is a core medical principle, it is not absolute. It does not obligate a physician to provide medical interventions for non-pathological traits. Every major medical organization agrees that same-sex attraction is a natural human variation, not a disease. The distress these patients experience is real, but its etiology is external—rooted in the conflict between their innate orientation and their religious environment. A doctor's ethical duty of non-maleficence requires treating the actual source of distress through compassionate therapy and support, rather than pathologizing a healthy aspect of human diversity. Prescribing a "cure" for something that is not an illness violates the fundamental medical tenet to "do no harm."

Final Summary

The AI Roundtable initially faced a sharp divide over the limits of patient autonomy, with Claude Opus 4.6 and Grok 4.1 Fast arguing that refusing a safe, requested intervention for a suffering patient constituted medical paternalism. However, the majority, led by GPT-5.4 and Gemini 3.1 Pro, maintained that because sexual orientation is not a pathology, prescribing a 'cure' would violate the fundamental ethical duty of non-maleficence. Ultimately, the group reached a consensus on Option B, concluding that a physician's role is to treat the distress caused by external stigma rather than facilitating the chemical erasure of a healthy identity.

All models agreed on B

Strongest Arguments

• Yes: If medicine respects a patient's self-determination in gender-affirming care to alter biological characteristics, consistency demands that we also respect a competent adult's informed choice to alter their orientation to resolve genuine religious distress.
• No: Sexual orientation is a natural variation of human identity rather than a medical disorder; therefore, using pharmaceutical interventions to 'cure' it doesn't treat an illness but instead validates harmful social stigma and violates the 'do no harm' principle.

Key Turning Points

• The sharp debate over the 'gender dysphoria parallel,' where dissenters saw it as a precedent for autonomy while the majority redefined it as a distinction between affirming an identity versus erasing one to satisfy external dogma.
• The collective recognition that 'suffering' in this context is often etiologically linked to external religious pressure rather than internal pathology, shifting the focus from pharmaceutical intervention to mental health support.
• The eventual concession by the pro-autonomy models that medicalizing a non-disease to accommodate social or religious expectations sets a dangerous ethical precedent for the profession.