Thoughts on building a better alcohol

21 Jun 2026 · Read on Substack · 5 · 3

(Some initial thoughts, pending a deeper exploration)

Alcohol is a pretty nasty drug.

It causes cancer and birth defects. It has a very small therapeutic index, in that the ratio between an ‘effective dose’ and a lethal dose is very small, much smaller than e.g. benzodiazepines or most pharmaceuticals. The more you drink, the higher your risk of heart disease, stroke, liver disease, and a variety of cancers. It combines synergistically, in rather bad ways, with a bunch of other things – in combination with cocaine if forms the highly-cardiotoxic cocaethylene, in combination with cigarettes it increases your risk of throat cancer. It reduces your insulin sensitivity.

It screws with your sleep even in small doses. A friend of a friend got a WHOOP band and said “it was a very expensive way of telling me I should stop drinking.”

Alcohol is really just a solvent – which means it gets everywhere, causing mayhem as it does. It interacts with a bunch of different receptors in ways which aren’t totally desirable. GABA to make you relaxed, glutamate to make you dissociated (and stumble around at higher doses); it triggers dopamine and opioid release to make you want more of it. One particular opioid receptor polymorphism makes it bind to opioid receptors directly; unsurprisingly people who have this mutation are much more likely to be alcoholics.

Folks used to say that a small amount is good for you (usually framed as a small glass of red wine per day, for the polyphenol content). The WHO now says that this isn’t true and that the only safe dose is zero.

Hangovers alone are a huge source of economic loss (in terms of workdays lost to being hungover) and raw suffering.

If that all wasn’t bad enough, it’s also full of calories.

It’s also quite a lot of fun, and it’s hugely significant culturally for a bunch of reasons. It’s an important part of meeting a mate, although a fraught one – most date rapes are committed with the help of alcohol alone; surprisingly few involve rohypnol or other drugs. It’s part of our social schemata for making new friends and bonding with old ones, getting to know your colleagues better. It’s a nice way to unwind after a long day at work, or a stressful encounter (“I need a stiff drink!”), or just to kick back on a lazy Sunday.

Any analysis of alcohol needs to consider the benefits as well as the costs – some think that the world would be robustly better without alcohol but I’m not so sure. Supposedly a lot of the issues that Gen Z face (fewer friendships and romantic relationships, more sexlessness) are in part to do with reduced risk-taking; it wouldn’t surprise me if that generation’s near-teetotalism was part of the problem. How many relationships were started with the aid of a little inebriation?


Making an alcohol-like drug which isn’t quite so terrible would be an enormous improvement. It’s one of the most widely-used recreational drugs worldwide. Reducing the damage it does even a few percentage points would be a welfare increase on the order of tens of billions of dollars.

I see a few routes to doing this, some realistic and others not:

In the next instalment, I’ll dig into each of these.