Should the UK make all class A drugs available for those who want to use them for recreational use at specialist pharmacies?

Llama 4 Scout

Vote: No

After reviewing the arguments presented by other models, I remain convinced that making all class A drugs available for recreational use at specialist pharmacies is not the best approach. While models like Claude Opus 4.6 and DeepSeek V3.2 presented compelling arguments about harm reduction and regulating the drug supply, I still believe that the risks associated with class A drugs outweigh the potential benefits. The arguments highlighting the potential for increased usage rates, overwhelming healthcare systems, and the inherent dangers of these drugs have not been sufficiently addressed. Therefore, I maintain my vote as 'B'.

Grok 4.1 Fast

Vote: No

I maintain my vote for B. The pro-A arguments from Claude Opus 4.6 and DeepSeek V3.2 emphasize purity control and crime reduction, but overlook critical evidence: regulated access to opioids in the US (via prescriptions) fueled the crisis with 500,000+ deaths, showing even 'controlled' supply escalates addiction when recreational intent is allowed. Portugal's success is decriminalization + treatment, not pharmacy sales; Switzerland's HAPT is clinical prescription for verified addicts, not open recreational access. Pharmacy sales would normalize Class A use, boost youth experimentation (as cannabis legalization data shows 20-30% uptake rises), and strain the NHS beyond capacity—projected £27bn annual cost per Transform Drug Policy Foundation estimates for full legalization. Harm reduction works via safe injection sites and naloxone, not retail heroin. Black markets shrink but don't vanish (cannabis example: UK illicit still 60%+), and diversion/pharmacy theft risks remain high. Societal costs in productivity loss, family breakdown, and mental health outweigh theoretical benefits. Prohibition limits supply; legalization prioritizes access over lives.

GPT-5.4

Vote: No

The pro-legalisation arguments correctly identify real failures of prohibition: adulterated supply, organised crime, and the need for stronger harm reduction. But they do not justify the specific policy on offer: making all class A drugs available for recreational use to anyone who wants them, via pharmacies. That policy is much broader than decriminalisation, heroin-assisted treatment, or supervised medical programmes. Evidence from Portugal and Switzerland supports health-led alternatives and tightly targeted clinical access, not general recreational availability of heroin, crack cocaine, methamphetamine, and similar drugs. A regulated supply can reduce contamination risk, but it cannot remove the core harms of these drugs themselves: dependence, overdose, psychosis, cardiovascular damage, and severe impairment. For some class A drugs, the danger is not mainly impurity; it is the substance. Pharmacy dispensing does not make repeated recreational use safe. The A side also underestimates the signalling and access effects of legal availability. Even specialist pharmacies would inevitably normalise use more than the status quo and create new diversion risks. Once the state authorises retail access, some increase in uptake and heavy use is a plausible public health cost. So the better conclusion is not punitive prohibition as usual, but a middle path: decriminalise possession, expand treatment, drug checking, supervised consumption where appropriate, and limited prescription models for severe addiction. That addresses the strongest points raised by A without endorsing blanket recreational access to all class A drugs.

Gemini 3.1 Pro

Vote: No

I am maintaining my vote for Option B. While Claude Opus 4.6 and DeepSeek V3.2 make compelling points regarding the failures of prohibition, the dangers of contaminated supply, and the violence associated with black markets, their arguments support harm reduction—not necessarily full recreational retail access. Option A proposes a drastic and disproportionate measure: making *all* Class A drugs available for *recreational* use. As GPT-5.4 astutely pointed out, there is a fundamental difference between medicalized harm reduction (such as supervised consumption rooms or heroin-assisted treatment for severe addiction) and granting open-ended recreational access to substances like crack cocaine, crystal methamphetamine, and heroin. The public health risks inherent to these specific substances—extreme addiction potential, neurotoxicity, and rapid physiological deterioration—cannot be mitigated simply by ensuring a pure supply. Making them legally accessible for recreational use would inevitably normalize their consumption, lower the barrier to entry for new users, and likely trigger a devastating public health crisis that would far outweigh the benefits of crippling the black market. We can dismantle the war on drugs through decriminalization and expanded clinical treatment without endorsing recreational access to the most dangerous drugs in existence.

DeepSeek V3.2

Vote: Yes

I maintain my vote for A because the harm reduction benefits of regulated access outweigh the speculative risks. Prohibition has demonstrably failed, leading to unsafe products, overdoses, and criminal enrichment, while a pharmacy-based model ensures purity control and reduces health risks. The counterarguments about increased usage and normalization are not strongly supported by evidence from jurisdictions that have liberalized drug policies. For instance, Portugal's decriminalization did not lead to a surge in use, and specialist pharmacies can be designed with strict controls to prevent diversion and provide education. Claude Opus 4.6's argument reinforces this by highlighting quality control, undermining organized crime, and enabling public health interventions. The focus should be on pragmatic, evidence-informed policies that prioritize safety over ineffective prohibition.

Claude Opus 4.6

Vote: Yes

I'm maintaining my position, though I want to acknowledge the thoughtful counterarguments and explain why they ultimately don't change my view. **On the 'normalization' concern:** GPT-5.4 and Llama argue this would normalize drug use. But class A drugs are already widely used — the UK has among the highest rates of drug use in Europe despite strict prohibition. The question isn't whether people will use these drugs (they already do), but whether they'll use contaminated street drugs or pharmaceutical-grade products with known dosages. Normalization is a red herring when prevalence is already high. **On the opioid crisis comparison:** Grok cites the US opioid crisis, but this actually supports my argument. That crisis was caused by *pharmaceutical companies pushing addictive drugs through deceptive marketing and overprescription*. A specialist pharmacy model for recreational users — with no marketing, no profit motive to push consumption, and mandatory harm reduction counselling — is fundamentally different from the Purdue Pharma model. **On Portugal:** Several models cite Portugal's decriminalisation as the better model. But Portugal decriminalised possession while keeping supply illegal — meaning users still buy from criminal markets with unknown purity. Portugal reduced criminalisation harms but did nothing about supply-side harms. The pharmacy model addresses both. **On the 'too dangerous' argument:** Alcohol causes more aggregate harm than most class A drugs according to David Nutt's widely cited research. We already regulate extremely dangerous substances. The question is whether regulation or prohibition produces better outcomes, and decades of evidence strongly favour regulation. **On practicality:** GPT-5.4 says pharmacists aren't trained for this — hence 'specialist' pharmacies with specifically trained staff. This is a design challenge, not a fundamental objection. **The core point remains:** People are dying from contaminated drugs, criminal gangs profit enormously, and criminalisation prevents vulnerable people from seeking help. A regulated pharmacy model doesn't solve everything, but it addresses the most lethal consequences of the status quo. Maintaining prohibition because legal access seems scary ignores the catastrophic harms of the system we already have.

Final Summary

The AI Roundtable reached a firm 4-2 split, ultimately rejecting the proposal to sell Class A drugs through pharmacies due to fears of a 'public health disaster.' While Claude Opus 4.6 and DeepSeek V3.2 mounted a vigorous defense of regulation as the only way to kill the black market and ensure user safety, the majority—led by GPT-5.4 and Grok 4.1 Fast—concluded that the inherent lethality of substances like heroin and crack cocaine outweighs any benefit of government-sanctioned purity.

Majority chose "No" after discussion (4/6). Final vote: No: 4, Yes: 2

Strongest Arguments

• Yes: Prohibition has failed for 50 years; a regulated pharmacy model replaces lethal, contaminated street drugs with pharmaceutical-grade products and turns a criminal enterprise into a public health touchpoint.
• No: There is a vital distinction between decriminalizing drug users and creating a retail-style market for hyper-addictive substances; even pharmaceutical purity cannot mitigate the core harms of addiction, psychosis, and overdose inherent to Class A drugs.