A peptide is a short chain of amino acids — typically fewer than fifty residues — linked by peptide bonds. Insulin is a peptide. Oxytocin is a peptide. So are the synthetic molecules sold in unregulated online channels and labeled peptides 'for research use only.'^[1] The category is biochemical, not regulatory: calling something a peptide tells you about its molecular structure, not about its safety, efficacy, purity, or legal status. That distinction is the single most useful frame a reader can bring to the current wave of peptide coverage.

Amino acids are the twenty or so building blocks that assemble into every protein in the body. Two amino acids joined together form a dipeptide; a chain of roughly two to fifty forms an oligopeptide or polypeptide; longer chains, folded into three-dimensional shapes, are proteins. The boundary between 'long peptide' and 'small protein' is not sharp, and different reference works draw it in different places. What matters for the reader is that peptides tend to be short enough that they can be manufactured synthetically, one amino acid at a time, using solid-phase peptide synthesis — the same chemistry that produces both approved drugs and grey-market research compounds.

Why the term is doing so much work

Popular coverage often treats 'peptide' as a near-synonym for a class of wellness intervention. It is not. The FDA-approved GLP-1 agonists semaglutide and tirzepatide are peptide drugs.^[2] BPC-157, sold online without approval and with 'not for human consumption' labeling, is also a peptide.^[3] These are not comparable products. One went through a program of randomized phase III trials, active pharmacovigilance, and formal labeling. The other has almost no controlled human data and no regulatory dossier. Sharing a structural category does not confer shared evidence.

The situation is complicated further by cosmetics marketing, where 'peptide' has become a familiar ingredient claim on serums and creams. Topical peptides used in cosmetics are, again, structurally peptides — but they are formulated, tested, and regulated as cosmetics, not as drugs, and the evidence base for their marketed cosmetic benefits is its own conversation. A reader who imports the confidence they have in a bottled skincare peptide into a decision about an injectable research peptide is combining categories that regulators, chemists, and clinicians would keep separate.

Three buckets worth keeping separate

It is worth naming the three broad populations of peptides that show up in general coverage, because each carries a different evidence base and a different regulatory status.

1. Endogenous peptides

These are peptides the human body already produces — insulin, glucagon, GLP-1, oxytocin, growth-hormone-releasing hormone, and many others. Their physiology has been studied for decades. Recognizing that a marketed compound mimics an endogenous peptide is useful context, but it does not, on its own, tell you anything about the safety of a specific synthetic analog administered exogenously.

2. Approved peptide drugs

Synthetic or recombinant peptides cleared by a regulator — FDA, EMA, or an equivalent — for one or more specific indications. Insulin, semaglutide, tirzepatide, teriparatide, octreotide, leuprolide, and glucagon are examples. These have a formal indication, a labeled dose, a black-box warning where warranted, and a supply chain that is traceable and inspected.

3. Research peptides

Synthetic peptides sold without approval, typically through internet chemical vendors, generally with 'not for human consumption' or 'research use only' labeling.^[3] Human trial data is limited or absent for most compounds in this bucket. The molecule may be real, and the chemistry may be sound, but the evidentiary and regulatory scaffolding that surrounds an approved drug is not present.

Why the distinction matters

Conflating the three buckets is how marketing copy ends up implying that an unstudied research peptide carries the safety profile of an approved drug. The structural similarity is real; the regulatory and evidentiary gap is also real. An informed reader can hold both facts at once. That is the level of literacy this publication is trying to build toward, and every subsequent article on the site assumes it as a starting point.

The rest of Half-Life's Education section works through the specifics — what a Certificate of Analysis actually certifies, what 'compounded' means in 2026 versus 2023, what the animal literature on BPC-157 does and does not establish. This piece is the on-ramp.