Your gut makes a hormone that tells your brain you have eaten, but your body destroys it in about two minutes. Semaglutide is a lab-made copy of that hormone, rebuilt to survive roughly a week, which is why it is given as one shot a week. People lose weight on it because they want less food, not because it blocks calories or burns more.
What it is supposed to do
It acts in three places at once. In the pancreas it helps release insulin, but only when blood sugar is actually high. In the brain it turns down appetite, and in the stomach it slows how fast food leaves. The weight loss comes mostly from the appetite part.
What people take it for
Most people take it to lose weight, and it is also an approved diabetes medicine. Some take very small weekly doses as an ongoing maintenance approach, which is a use no trial has tested.
The dose question
What circulates
Prescription doses, often from grey market vials
What was studied
Large randomised placebo controlled trials
The strongest evidence base of anything on this site
The risk moved off the molecule and into the vial. Three tested vials were far less pure than claimed and simultaneously over label, so careful dosing meant overdosing.
Reported because it is what people use. Nothing here recommends any amount.
How long it lasts
About seven days in people, which is why it is a weekly shot. The natural gut hormone it is copied from lasts about two minutes.
Route studied
Weekly injection of pharmaceutical product, in large trials. Also approved as a pill.
Route used
Weekly injection at home, often mixed by the person from a powder vial.
How to check you have the right molecule
The only check that means anything is an independent lab test on the actual batch in your hand, compared against two numbers. When three vials bought from online sellers were tested in 2024, purity came back at 14.37%, 8.97% and 7.7% against the 99% advertised, meaning the powder was mostly other material. At the same time all three held 28.6 to 38.7 percent more actual semaglutide than the label claimed. All three carried bacterial toxin, and against 22 standard pharmacy quality criteria the samples failed 13 and 14.
People assume the huge trial results and the number on the label both apply to the vial that arrived in the mail. No trial has ever used research grade or compounded material, and the one published test of grey market vials found the street belief backwards: the product was not weak and underdosed, it was dirty and stronger than the label said, which is how somebody dosing by that label overdoses.
What is known about harm
Gut trouble is the common one. In one trial 63.5% of people on the drug reported nausea, diarrhea or similar, mostly mild, and 4.5% quit over it against 0.8% on placebo. A study of 4,757 new users found pancreatitis at roughly nine times the risk, though its own margin of error runs from barely raised to 66 times, so the real figure is genuinely uncertain, plus bowel obstruction about four times and a stomach that stops emptying properly about 3.7 times. The label carries a boxed warning about thyroid tumors based on rats, whether that carries over to people is undetermined, and nobody knows what ten years of continuous use does because the longest randomized data is two years.
Everything on this page is a summary. The citations, the studies and the reasoning live in the full entry.