When you think that of tools for studying substance use and addiction, a social media site like Reddit, TikTok, or YouTube probably isn't the very first thing that involves mind. But the stories shared on social media platforms offer unprecedented insight into the world of substance use.
In the past, researchers studying people's experiences with addiction relied totally on clinical observations and self-reported surveys. But only about 5% of individuals were diagnosed with a substance use disorder seek formal treatment. They make up only a small proportion of the population affected by a substance use disorder – and until recently there was no easy option to capture the experiences of the opposite 95%.
Today, tens of millions of individuals openly discuss their experiences with drugs online, creating an enormous collection of raw narratives about drug use. As Doctoral student in information science With a background in public health, I take advantage of this material to higher understand how individuals who use drugs describe their lives and understand their experiences, particularly in terms of stigma.
These online conversations are changing the way in which researchers take into consideration substance use, addiction and recovery. Advances in artificial intelligence are helping to know these conversations to a level that was impossible before.
The hidden population
The overwhelming majority of individuals diagnosed with a substance use disorder address the issue informally, searching for support from their community, friends, or family. Self-medicate or do nothing in any respect. However, some decide to post about their drug use in dedicated online communities comparable to group forums, often with a level of openness that might be difficult to capture in clinical interviews.
Her social media posts provide a glimpse into real-time, unscripted conversations about substance use. For example, Reddit, which consists of thematic communities called subreddits, comprises over 150 interconnected communities Communities dedicated to varied elements of substance use.
In 2024, my colleagues and I analyzed how participants network and interact in drug-related forums on Reddit. We found that they focused on the chemistry and pharmacology of drugs, supporting drug users, recreational experiences comparable to festivals and book clubs, aiding in recovery, and harm reduction strategies. We then chosen among the most energetic communities to develop a system for categorizing several types of personal disclosures Labeling 500 Reddit posts.
mucahiddin/iStock via Getty Images
Policymakers and health experts have raised concerns about this Social media promotes dangerous drug use. Our work didn't examine this issue, nevertheless it did support the concept that platforms like Reddit and TikTok often function a lifeline for people searching for just-in-time support after they need it most.
When we used machine learning to investigate an extra 1,000 posts, we found that almost all users were searching for practical security information within the forums we focused on. The posters often asked questions comparable to how much of a substance is protected to take, what interactions ought to be avoided, and find out how to recognize signs of discomfort.
We have observed that these forums function as informal harm reduction spaces. People share not only experiences, but additionally warnings, safety protocols, and real concern for one another's well-being. When community members are lost to an overdose, the response shows deep sadness and renewed commitment to making sure the security of others. This is the on a regular basis reality of how people take care of substance use outside of medical settings – with way more nuance and mutual support than critics would expect.
We also explored TikTok, Analysis of greater than 350 videos from substance communities. Content promoting recovery was probably the most common, appearing in 33.9% of the videos we analyzed. Only 6.5% of the videos showed energetic drug use. Like Reddit, we frequently saw individuals who valued safety and care.
Why AI is changing the sport
Platforms like Reddit, TikTok and YouTube host tens of millions of posts, videos and comments, many stuffed with slang, sarcasm, regional language or emotionally charged stories. Manual evaluation of this content is time-consuming, inconsistent, and virtually inconceivable to perform at scale.
This is where AI comes into play. Traditional machine learning approaches often depend on fixed word lists or keyword matching, which might miss vital contextual clues. In contrast, newer models like this – especially large language models GPT-5 by OpenAI – are capable of understand the nuances, tone and even the underlying intent of a message. This makes them particularly useful for studying complex topics comparable to drug use or stigma, where people often communicate through innuendo, coded language, or emotional nuance somewhat than direct statements.
These models can discover patterns across hundreds of posts and discover emerging trends. For example, researchers used them to detect changes in the way in which Canadians interact on X, the social media site formerly called Twitter. discussed cannabis as legalization approached – Capture changes in public attitudes that traditional surveys might need missed.
In one other study, researchers found that monitoring Reddit discussions will be helpful Predict opioid-related overdose rates. Official government data, comparable to that from the Centers for Disease Control and Prevention, typically lags behind by no less than six months. But adding near-real-time Reddit data to predictive models significantly improved their ability to predict overdose deaths — which could potentially help public health officials respond more quickly to emerging crises.
Bringing stigma into focus
One of probably the most difficult elements of substance use to check and address is stigma.
It is deeply personal, often invisible, and shaped by an individual's identity, relationships, and environment. Researchers have long recognized that stigma, particularly when internalized, can undermine self-esteem and worsen mental health and stop people from searching for help. However, they’re notoriously difficult to capture using traditional research methods.
Most clinical studies are based on surveys or interviews conducted at regular intervals. While these snapshots are useful, they will overlook how stigma spreads into on a regular basis life. Stigma researchers have emphasized the necessity to know its full impact Pay attention to how people speak about themselves and their experiences over time.
On social media platforms, people often discuss stigma organically, in their very own words and within the context of their lived experiences. They could describe being assessed by a healthcare providerexpress shame about their very own substance use or take into consideration how stigmatization shapes their relationships. Even when contributions do indirectly name the experience as stigma, they still show how stigma is internalized, challenged, or reinforced.
Using large language models, researchers can begin to trace these patterns at scale and discover linguistic signals comparable to shame, guilt, or expressions of hopelessness. In more moderen worksMy colleagues and I even have shown that the stigma expressed on Reddit aligns closely with long-standing stigma theory—suggesting that what people share on social media reflects recognizable stigmatization processes and is just not something fundamentally latest or different from what researchers have long studied.
This is essential because stigma is certainly one of the most important barriers to treatment for individuals with substance use disorders. Understanding how individuals who use drugs speak about stigma, harm, recovery and survival in their very own words can complement surveys and clinical studies and help educate higher public health responses.
By taking these on a regular basis expressions seriously, researchers, clinicians, and policymakers can begin to reply to substance use because it is definitely lived—messy, evolving, and deeply human.
