Technology

TikTok Blocks Peptide Searches, Not the Videos Behind Them

Martin HollowayPublished 60m ago3 min readBased on 4 sources
Reading level
TikTok Blocks Peptide Searches, Not the Videos Behind Them
Photo by N509FZ / CC BY-SA 4.0

TikTok searches for semaglutide, BPC-157 and retratutide return a "no results found" page. The block was in place as of Oct. 20, 2026, and applies to exact queries for those compounds in TikTok search The Verge.

Searches for BPC-157 and TB-500, the ingredients in the so-called "Wolverine" peptide stack, are also not searchable. The limit works at the query level. It makes those exact terms harder to discover without deleting the underlying videos.

The general term "peptides" still returns many short videos. TikTok has not removed existing peptide videos while blocking certain search terms. Slight variations in spelling of the blocked names still return published videos.

In practical terms, this kind of exact-match filter lowers visibility for casual search but leaves other paths open. Direct links, followed accounts, reused audio and alternate spellings still function. It changes what search returns. It does not erase the library behind it.

TikTok has used similar limits on the same search box before. It began restricting accounts posting about weight-loss drugs such as Ozempic and Wegovy in 2023 Bloomberg. In 2024, reporting found users experiencing nausea and constipation were turning to TikTok creators for advice about Ozempic side effects Bloomberg. Those Ozempic creators later said the new rules singled them out unfairly Bloomberg.

The broader context here is a choice between search governance and content governance. Blocking a small set of high-intent queries, meaning searches that show a clear plan to find a drug, is fast to ship, simple to audit and easy to reverse. It also degrades quickly. Creators and viewers adapt with spacing, sound-alike words, abbreviations and adjacent hashtags. Moderation teams must then decide whether to expand the blocklist or accept that some circulation continues.

The practical concern for teams trying to measure this is that the current state looks like partial containment rather than removal. The video corpus, the full collection of stored videos, persists. Discovery continues under broader or disguised queries. Checking only the blocked strings will overstate coverage. A complete check has to track variant queries, recommendations and inbound traffic from outside search.

In my view, that partial step still makes sense as a first control, if it is understood as friction and not as enforcement. Search friction can reduce accidental exposure for someone typing a drug name cold, while policy, clinical review and tooling teams take time to decide what, if anything, should be labeled or removed. The risk is reading a quiet search page as a solved problem. The earlier history with Ozempic and Wegovy suggests demand for peer anecdotes about dosing, side effects and sourcing does not disappear when a query goes dark. It migrates.

Looking longer term, the hopeful possibility is better instrumentation, meaning better tracking of search behavior. Query blocks create clear signals about attempted access, which workarounds emerge and where repeat evasion happens. Used well, those signals can support clearer rules, more consistent review and pages that direct high-risk queries to authoritative sources instead of an empty result. The technology for that exists. The hard part is still deciding which queries qualify and keeping the list current when users push back.