Insulin is a peptide. So is semaglutide. There are more than a hundred FDA-approved peptide drugs in the United States, and their safety profiles are documented in detail.
BPC-157 bought from a website is also a peptide, and almost nothing about it is documented at all.
One word covers both, which is why the question has no single answer. There is a better question underneath it, and that one does: what actually goes wrong, and how often.
What the word is doing
Four very different things get called peptides, and they sit in different places on safety.
Approved medicines come first. Insulin, semaglutide, tirzepatide, teriparatide and around a hundred others, tested in thousands of people, watched after approval, with side effects printed on a label you can read.
Collagen and food peptides are protein fragments you swallow. A six-month trial for knee arthritis found no serious side effects, and the worst most people get is an upset stomach at high doses.1
Then the grey market. BPC-157, TB-500, CJC-1295, ipamorelin, retatrutide and the rest of this library, sold for lab use, injected by people anyway, with almost no trial data in people.1
And investigational compounds, in trials right now, with rules, monitoring, and a doctor watching.
Almost everyone asking this question means the third group. The honest answers get complicated there, and they stop being about chemistry.
What actually goes wrong
What is actually in the vial
Researchers bought semaglutide from illegal online pharmacies and tested it in a laboratory. The results were not close.
Labels claimed 99% purity. Measured purity came in between 7.7% and 14.37%. The amount of semaglutide present overshot the labelled amount by 28% to 39%. And endotoxin, a bacterial contaminant that causes fever and inflammatory reactions, was detected in every single sample, at levels between 2.16 and 8.95 endotoxin units per milligram.2
Less than a seventh as pure as claimed, more drug present than the label says, and bacteria in all of it. Anyone following that label was dosing off a number with no relationship to what was in front of them.
Published case reports show what that looks like in a person. One documents euglycemic ketoacidosis, a serious metabolic emergency, following the use of counterfeit semaglutide for weight loss.3 Another found a liraglutide formulation compromised by hidden impurities in common pharmaceutical ingredients.4
In July 2026 the head of a peptide company was sentenced to 70 months in prison for selling tainted and substandard products.6
And the scale is not small. US peptide imports from China reached $328 million in the first nine months of 2025, close to double the same period the year before. LegitScript recorded a 308% increase in problematic peptide advertisements between 2023 and 2024, and a 678% increase against 2022.2 Roughly 100 million prescriptions were dispensed through illegal online pharmacies in 2023, about half again as many as in 2019.5
Nothing in any of that is about what a peptide does in a body.
Dosing errors
The second failure, and the one people brush off because it feels like a test of their own competence rather than a risk.
Someone went through 112,532 FDA adverse event reports and found that 63% of the GLP-1 reports were about how the drug was given, against 39% for insulin.8 Same route, same kind of syringe, and far more of the trouble coming from the giving rather than the drug.
Those errors began climbing in late 2022 and kept rising through 2023 and 2024. Insulin showed no such pattern. The timing matches the GLP-1 shortages, when more people started getting these drugs from places that came with less instruction.8 The FDA put out its own alert about people accidentally giving themselves more than prescribed after getting the draw wrong from a multi-dose vial.8
Injection technique
Newer, and it fills the gap between the other two.
In August 2026, researchers watched people who had bought peptides online actually inject them. Several could not clear the large air bubbles from the syringe, which throws the dose off and irritates the site. Some injected into muscle instead of the fat on the arm or belly, which changes how fast it gets absorbed.9
Nobody had taught them. That absence is the finding. An approved drug arrives with a device built to stop exactly these mistakes and a pharmacist to explain it. A vial from a website arrives with neither.
What is documented for approved peptides
The answer here is reassuring, and it gets lost in the noise around everything else.
For GLP-1 drugs, the common problems are in your stomach. Nausea turns up in 2% to 20% of patients depending on the study, and about 10% of people stop because of side effects.1
The thyroid cancer worry needs an update. It came from studies in rats and produced a warning that shaped a decade of public fear about this whole drug family. A 2026 review of the human evidence found no rise in thyroid cancer among people taking semaglutide, at under 1% and 0.001 cases per 100 patient-years.1 The warning is still on the label. The human evidence does not support the fear it created.
What is unknown for the rest
Not a gap in this guide. A gap in the world, and it has a shape.
Most compounded and grey-market peptides have never been through the trials that settle a dose, a side effect list, or who it helps.10 There is no long-term data, so questions about what months or years of use does to your organs have no answer. Quality swings between suppliers, because compounding is spread across thousands of pharmacies and research vendors are not regulated at all.
And there is a quieter problem underneath all of that. When somebody is harmed by a compounded or grey-market peptide, it often never gets reported.10 So a quiet safety record is not proof that nothing is happening. It may just mean nobody counted.
A watchdog review in 2026 listed serious problems tied to some peptides, including heart events, kidney trouble, melanoma and infections, and noted that the companies selling them are not collecting or reporting any of it.11
The rules are being loosened while the evidence stays flat
Something is happening in the background that anyone reading about peptide safety should know.
In 2023 the FDA judged 19 peptides too unsafe for compounding pharmacies to dispense. The agency is now positioned to reverse that. Little new science has appeared on those 19 since 2023, while demand has climbed steeply on the back of social media promises about physique, skin, hair, injury recovery and energy.7
A former acting FDA commissioner described the potential reversal as a disruption of the arrangement in place since 1962, under which drugs are studied to see whether they work before being marketed. Officials have warned that justifying the reversal on the grounds that there were no safety concerns would lend a false appearance of safety to more than a dozen unapproved, untested drugs.7
Take from that what you like about the policy. The relevant point for safety is narrower: rules changing is not evidence changing. A compound becoming easier to obtain tells you nothing about what it does in a body over ten years, and the loosening is happening without the studies that would answer that.
So what is the actual answer
Three statements, all true at once.
Approved peptide drugs have documented safety records and are about as well understood as any medicine.
Grey-market peptides mostly have no safety record at all, and the honest position on their long-term effects is that nobody knows.
And the things that actually put people in a doctor's office here are not the exotic ones. They are a dirty vial, a decimal point, and an air bubble.
Anyone can do something about two of those three tonight, with no medical training at all. More useful than a verdict on the whole category.
Common questions
Are peptides safe?
Depends entirely which ones. More than a hundred are FDA-approved with documented safety records, including insulin. Peptides bought from research vendors have almost no human safety data, and the problems that get reported tend to come from contamination or a wrong dose rather than from the compound.
What is the most common thing that goes wrong?
Contamination and wrong doses, and neither is about the molecule. When researchers tested semaglutide bought from illegal online pharmacies, measured purity ran between 7.7% and 14.37% against a claimed 99%, and bacterial endotoxin was present in every sample.
Do peptides cause cancer?
For semaglutide, a 2026 review found no rise in thyroid cancer, at under 1% and 0.001 cases per 100 patient-years, despite a warning that came from rat studies. For most research peptides there is no long-term human data either way, which is not the same as a clean bill of health.
Are the side effects worse than steroids?
The comparison does not hold. They are unrelated, and most research peptides have no side effect list to compare against. Approved peptide drugs generally have mild, well-documented side effects. Unapproved ones have unknown ones.
If it is sold legally, does that mean it passed a safety check?
No. Research vendors sell under a lab-use label, and that requires no safety testing of any kind. Legal to sell and checked for safety are two different things.
Why do people not report problems?
Reporting for compounded and grey-market products is patchy. No rule says a research vendor has to collect reports, and someone who bought something labelled not for human use often does not think to file one. So problems here are much harder to spot than in approved medicine.
The FDA is loosening the rules. Does that mean these are safe now?
No. Little new science has appeared on the 19 peptides restricted in 2023, and a former acting FDA commissioner has described the reversal as a break from the arrangement in place since 1962 that drugs are studied before being marketed. Rules changing is not evidence changing.
What is actually worth watching for?
The things that actually go wrong rather than the exotic ones. Redness, heat or pus at an injection site points to infection. A fever after injecting is worth taking seriously. And a dose that came out very different from what you expected is usually a maths error rather than a bad reaction.
References
- [Are peptides bad for you? Risks and side effects](https://medicalnewstoday.us/are-peptides-bad-for-you/), 2026. *Review of published literature. Separates approved, collagen, grey-market and investigational categories. Cites a randomised trial of collagen peptides reporting no serious adverse events over six months; GLP-1 nausea rates of 2 to 20% with roughly 10% discontinuation; and a 2026 systematic review in Diabetes, Obesity and Metabolism finding no increased thyroid cancer incidence among semaglutide-treated patients.* ↩
- [Research-grade peptide contamination: the risks of unregulated products](https://rethinkpeptides.com/articles/research-grade-peptide-contamination-the-risks-of-unregulated-products), 2026, citing the *Journal of Medical Internet Research*, November 2024. *Laboratory analysis of semaglutide purchased from illegal online pharmacies. Measured purity 7.7% to 14.37% against 99% claimed on labels; semaglutide content exceeding labelled amounts by 28.56% to 38.69%; endotoxin detected in all samples at 2.1645 to 8.9511 EU/mg. Also reports US peptide imports from China of $328 million in the first nine months of 2025, and LegitScript figures of a 308% rise in problematic peptide advertisements between 2023 and 2024.* ↩
- Sterckx M, et al. Euglycemic ketoacidosis following the use of counterfeit semaglutide for weight loss. *Cureus*, 2026. *Case report.* ↩
- Sheikh AR, et al. Liraglutide formulation can be compromised by hidden impurities in common pharmaceutical ingredients. *Journal of Pharmaceutical Sciences*, 2024. *Laboratory analysis.* ↩
- IQVIA Institute, Use of Medicines in the US, 2024, as cited in the FAERS analysis at reference 8. *Estimates nearly 100 million prescriptions dispensed through illegal online pharmacies in 2023, close to 50% more than in 2019.* ↩
- [DIY peptide injections are risky](https://medicalxpress.com/news/2026-08-diy-peptide-risky.html), 2026. *Written by a pharmacist and clinical pharmacologist studying the grey market. Describes contamination risk, purity shortfalls and dose inaccuracy, and the July 2026 sentencing of a peptide company chief executive to 70 months for selling tainted and substandard products. Also reports the Public Health Reports injection-practice study.* ↩
- [Peptide safety, the FDA and compounding pharmacies](https://www.propublica.org/article/peptide-safety-fda-compounding-pharmacies), ProPublica, 2026. *Investigative reporting. Covers the potential reversal of the 2023 decision restricting 19 peptides, the absence of new science since, and the warning from a former acting FDA commissioner about the 1962 arrangement that drugs be studied before marketing.* ↩
- [Adverse events administering GLP-1 receptor agonists: a cross-sectional study](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12927500/). *Analysis of 112,532 FDA Adverse Event Reporting System reports, January 2015 to December 2024. GLP-1s associated with 63% administration-related reactions against 39% for injectable insulin. Dosing and administration errors rose from Q4 2022, a pattern insulin did not show. Notes an FDA alert on patients inadvertently administering higher doses than prescribed after errors drawing from multi-dose vials.* ↩
- Study of grey-market peptide injection practice, *Public Health Reports*, August 2026, reported in [DIY peptide injections are risky](https://medicalxpress.com/news/2026-08-diy-peptide-risky.html). *Observational. Found users struggling to clear large air bubbles and injecting into muscle rather than subcutaneous fat.* ↩
- [Why peptide compounding rules matter in 2026](https://centreformedicalhumanities.org/fda-peptide-deregulation-2026-health-news-academic-review-2026/), 2026. *Academic review. Notes the absence of phase 1 to 3 trial data for most compounded peptides, quality variability across decentralised compounding, absence of long-term safety data, and spotty adverse event reporting that makes emerging safety signals hard to detect.* ↩
- [Wellness peptides lack reliable safety information, watchdog warns](https://www.medscape.com/viewarticle/wellness-peptides-lack-reliable-safety-information-watchdog-2026a1000fbf), Medscape, 2026. *Reporting on a watchdog review. Lists serious adverse events associated with some peptides including cardiac events, kidney dysfunction, melanoma and infections, and notes that companies selling them are not systematically collecting this information.* ↩