We’re Not Ready For the New Era of Drugs
AI drug discovery, misinformation, and loneliness are set to rip us apart.
There’s something so magnetic about research chemicals. The idea that drugs exist which are not only rare but weird, has always been attention-grabbing. Some people might approach them with a sense of smugness, thinking they’re above their peers because their substance is harder to get hold of. Others might enter the space driven primarily by curiosity.
Even their nickname, designer drugs, has a ring to it. However, these substances typically fail to live up to that nickname– they might not be mass-produced, but they aren’t made only for the wealthy and famous, nor are they particularly expensive.
Research chemicals somehow occupy the underside of a reality that is already within the underside. Substance-use culture is, for the most part, the counterculture, and yet research chemicals have dug deeper to become the counter-counterculture.
Whatever commonplace substance you can think of, there’s a research chemical out there which performs its role, albeit to a stranger and lesser-known degree. The research chemical community is also unusual. They’re scientific, curious, reckless, cunning, and by far the smartest people I’ve ever encountered.
Some of this is because of their own desire to try out and uncover the edges of the drug world. But much of it comes from the fact both finding these substances and finding information about them is hard. Research chemicals are relegated to the sidelines because they never found their footing societally, meaning they’re an acquired taste even by those who like getting high. There’s less market demand for them, and so there are fewer suppliers.
Many labs don’t want to put themselves at risk by making substances that’ll only be enjoyed by a niche few, as the returns simply aren’t there. Therefore, if you’re interested in research chemicals you need to really learn how to track them down. What’s more, if you manage to get your substances, you need to read up on safe usage as much as possible.
Let’s say you’re interested in dissociative research chemicals. There’s a vast range out there. 3-MeO-PCE, 3-MeO-PCP, OPCE, 2FDCK, MXE. The list goes on. Some of these are easier to read up on than others. If you find yourself in especially unusual territory, then you might get your hands on a research chemical that only has a couple of posts online describing its function, feeling, and dosage.
I’ve highlighted two problems here:
- Research chemicals are hard to find
- Information about research chemicals is also hard to find.
These are both complications, although for the most part they’ve weirdly balanced each other out. The more popular a research chemical is, the more likely you’ll find information and guides about them. Typically, it’s only the really obscure ones that suffer the worst fate. However, I predict the age we’re about to enter is going to favour obscurity so much more than it already does. And I don’t think we’re prepared for what comes next.
AI Drug-Discovery
Drug discovery is one of the largest sectors of the AI world. There’s a lot of money going into machine learning and substance creation. Forecasts indicate this market will be worth $2.65 billion USD by the end of 2025. Google echoes this sentiment, announcing a $20 million fund to support its AI protein-folding service, AlphaFold.
While this doesn’t generate the same eye-catching headlines as large language models do, they occupy a meaningful part of the industry. Most AI drug-discovery tools are being used either by pharmaceutical companies or research institutes. At the moment, they’re directed towards trying to solve serious and specific medical issues, or are being experimented with to test their capabilities. But this way of doing things tends to trickle into the recreational world.
The traditional pipeline for synthetic recreational drugs is that they start off as discoveries made by researchers, who either abandon them or try to study their benefits. Then, after some time, they’re discovered to have recreational properties and so they seep into the market. Researchers discovered an enormous set of potential psychedelics last year, and we could soon see AI drugs following a similar path imminently.
While this could usher in an exciting new time for the substance-use world, with new drugs flooding the market and offering more variety, I’m concerned about the downside– when more drugs come out, it’s much harder to get information about them. They get documented less often, and those who do document them might struggle to get the word out as topics like this tend to get hidden by Google. We already have blindspots in the research chemical world, but this could make it so much worse.
You might read this and think that it shouldn’t really matter, because the most obscure substances will cause the least harm as fewer people will have them. I made a similar argument earlier, so I don’t disagree. But my point is if AI drug discovery becomes commonplace, and it becomes easy to develop these substances, then we might find that the amount of new drugs on the market means hardly any of them develop popularity, and so most stay obscure. This somewhat already happened with the synthetic cannabis sector, as variations were cheap to make and deeply varied, but because they were all within one category of drug, it was easier to mitigate by giving generalized advice. The world of research chemicals is too vast for that to happen.
The caveat
Just because drugs are getting discovered doesn’t mean we’ll figure out how to make them, or that they’ll be easy enough to make in an underground lab. So even if there are thousands of new drugs are discovered, they may never actually extend into reality. Having a recipe is one thing– having the ingredients and equipment is another.
But I wouldn’t take this as a salve to the problems ahead. AI drug discovery is in an adolescent phase, like most of the AI market, but that could change soon. In the future, these applications won’t simply display potential substances, they’ll likely be configured to each lab’s specific setups, supplies, and tools, so they only show possibilities that can actually be made. Therefore, even if a fraction of AI drugs can get produced, that fraction could still be meaningful.
The Age of Misinformation
The prospect of AI drug discovery is made much worse because information about drugs is already hard to come by, even for popular ones. Ever since the War on Drugs, there’s been serious misinformation (and disinformation) going around. Part of the campaign against substances has always been to restrict knowledge about them, which further endangers people.
The result is we’re somehow both less trustworthy and more trustworthy. We don’t listen to authority about drugs, but we might listen to a random TikTok, Reddit comment, or Medium article giving advice. This is especially dangerous when it comes to drugs that are so novel. When you can’t find out much about a drug, you’ll cling onto any story or graph or statistic you can find about it. And if AI drug discovery floods the market, then that might become even more of the norm.
We might not be able to combat that, but we can teach people basic facts about drugs so they can tell when something sounds fishy online. Maybe we can’t keep up with the sheer number of research chemicals out there, but we can inform people that mixing dissociatives with most depressants can lead to blacking out, or that benzodiazepines cannot be taken with alcohol for fear of death, or that psychedelics do not simply feel like stronger versions of weed. These are all facts that many newcomers to the drug-world might not know because of suppressed information.
This is a start, but it won’t be enough. We probably need advice that’s dialled more into the research chemical space. For instance, a PCP analogue will probably give you energy as well as dissociation is a common one that isn’t so easy to find out straight away. Teaching how to reagent test is important, too. Each substance might react differently, but if we can build new drugs, we can build new testing methods, too.
Loneliness Adds Fuel to the Flames
It seems the most lingering effect of the COVID pandemic is its impact on mental health. Loneliness is on the rise, and we seem unable to control it. The world looks back to normal, but there are stark changes that cannot be reversed so quickly. Some people cut away from friends and colleagues and never reconnected. Some people were so young during the pandemic they missed out on formative lessons from school around socializing and forming bonds. Some people developed mental illnesses that contributed to their isolation.
When people feel lonely, drugs can act as a panacea. Maybe somewhere we’ll discover a medical cure for loneliness (if that even makes logical and phenomenological sense), but regardless, we’re sure to create some substances that disconnected characters gravitate towards. It’s not that drugs are only taken alone– it’s that without others around, there isn’t anybody to hold you back, and there’s nobody to save you if you overdose.
We need greater acceptance of drug-use, so that people don’t hide their activities. We need rave-culture to return properly, so that people can meet with like-minded individuals. We need online spaces that work hard to offer real education, and perhaps even companionship. Maybe medical tech can help out here, too. Perhaps some bracelet or ring can monitor your vitals and alert emergency services if you show dangerous signs.
I’m not going to say any of these will be enough, but they could help. Loneliness is one of the harder problems to solve in the world, because unlike depression or anxiety, loneliness requires participation from multiple people. It’s not just a mental illness, it’s a description of a person’s physical surroundings. It’s also a description of our environments.
Final Words
None of the problems I’ve suggested are easy to solve. But they’re not impossible to fix either. More than anything, they require an active engagement with substance-use culture on a societal level, so people understand what’s happening, and are not suppressing helpful information. It also requires us to look out for each other, and build spaces that allow us to meet and get more connected.
Above all, note that this is a hypothetical. While I definitely see this as being a future we’re heading towards, it’s not set in stone by any means. We still have time, and it’s why those who can write about substances should. Their voices are needed now more than ever.
For further critical discussions on substances, culture, and medicine, check out my physical magazine, Existential Horror. This is a 250+ page publication designed to be a critique and celebration of psychedelia and substance-use culture. Vol 2 is out now!
