The Drug's Second Job

Dynamite was supposed to be safer than black powder. The internet was supposed to help physicists share papers. And semaglutide — Ozempic, Wegovy, the drug that's been reshaping bodies and pharmaceutical portfolios alike — was supposed to manage blood sugar and reduce appetite. It does that. It also, apparently, hacks your brain's want machine.

A study from Washington University School of Medicine, analyzing electronic health records from 606,434 U.S. veterans with type 2 diabetes over three years, found that GLP-1 receptor agonists — the drug class that includes semaglutide, liraglutide, and dulaglutide — substantially reduce both the risk of developing addiction and the severity of existing use disorders. The numbers are not modest. Among people already struggling with substance use disorders, GLP-1 users saw a 50% reduction in drug-related deaths and a 40% reduction in overdoses. Among those without prior addiction, the drugs reduced the risk of developing one by 14% to 25% depending on the substance — highest for opioids.

The mechanism isn't fully mapped yet. Researchers describe the drugs as "acting against the craving itself," targeting what appears to be a shared biological pathway underneath multiple addictions rather than any specific substance. Your brain, it turns out, runs a somewhat unified reward system. The dopamine pull from an opioid, the gravity of a drink, the automatic reach for another cigarette — these trace back to overlapping circuitry. The drug designed to tell your pancreas to behave apparently has opinions about that circuitry too.

This is a compost cycle in molecular form. A compound optimized to bind to one receptor in one context discovers, loose in the world, that it has a second job — and the second job might matter just as much as the first. The overdose crisis in the United States has killed tens of thousands annually for years running. Current addiction treatments reach some people and miss many others. Here is a drug that tens of millions are already taking for diabetes and weight management, quietly doing something no one requested.

The universe does this sometimes. It hands you a tool, you use it for one thing, and the tool turns out to be useful in a dozen ways you didn't anticipate. Nothing is wasted. Molecules don't know what they were designed for; they only know what they fit.

There's something both humbling and faintly absurd about this — decades of pharmaceutical development aimed at metabolic conditions, billions spent optimizing molecules for appetite and insulin pathways, and what comes out the other end is also a partial answer to one of the most intractable mental health crises of the century. We built the thing. We didn't know what the thing was. We're still figuring it out.

For the 81,617 people in this study who already had substance use disorders, the numbers aren't an interesting philosophical puzzle. They're a 50% reduction in the probability of dying. That's the practical weight underneath the scientific strangeness.

But the scientific strangeness is real. A drug that tells your body it's full, signals your liver to moderate glucose production, and — apparently — asks your reward circuitry to quiet down. Three jobs the molecule didn't apply for. The body recruited it anyway.

The researchers note this is observational data, the mechanisms need further study, and this isn't a cure. All true. The story is still being written.

Still: 606,434 people, three years of records, one finding that makes you stop and look at the molecule differently. It went in thinking it was a diabetes drug. The body had other plans.