pith. sign in

arxiv: 2508.16453 · v2 · pith:IP2JEXJUnew · submitted 2025-08-22 · 💻 cs.SI · cs.CL· cs.LG

Anti-establishment sentiment on TikTok: Implications for understanding influence(rs) and expertise on social media

Pith reviewed 2026-05-21 22:51 UTC · model grok-4.3

classification 💻 cs.SI cs.CLcs.LG
keywords anti-establishment sentimentTikToksocial mediaconspiracy theorieswellnessfinanceengagement patternsinfluencers
0
0 comments X

The pith

Anti-establishment sentiment is common in TikTok conspiracy content but rare in finance and wellness videos.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper examines the prevalence of anti-establishment sentiment on TikTok in three areas where creators often claim expertise. It applies a computational method to detect distrust of institutions in posts about finance, wellness, and conspiracy theories. The analysis shows this sentiment appears most often in conspiracy videos and infrequently in the other two topics. Engagement with such posts differs across areas, which points to possible platform incentives that reward distrustful framing. Understanding these patterns matters for seeing how social media shapes attitudes toward public institutions.

Core claim

Anti-establishment sentiment is most prevalent in conspiracy theory content on TikTok and relatively rare in content related to finance and wellness. Engagement patterns with such content vary by area, and there may be platform incentives for users to post content that expresses anti-establishment sentiment.

What carries the argument

Computational labeling of TikTok posts for the presence or absence of anti-establishment sentiment, applied across finance, wellness, and conspiracy domains.

If this is right

  • Creators positioning themselves as experts in finance or wellness rarely use anti-establishment framing to gain attention.
  • Conspiracy content relies more heavily on distrust of institutions to attract viewers.
  • Differences in engagement suggest that anti-establishment posts can receive different levels of interaction depending on the topic.
  • Platform design choices may encourage certain types of content that question institutional authority.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • This pattern could affect how users evaluate health or money advice from non-institutional sources.
  • Similar labeling could be tested on other short-video platforms to check if incentives appear elsewhere.
  • The findings leave open whether repeated exposure to rare anti-establishment posts still shapes broader distrust over time.

Load-bearing premise

The computational method correctly identifies anti-establishment sentiment in posts without major errors or topic-specific biases.

What would settle it

A hand-coded sample of posts from the finance and wellness sets showing frequent mislabeling as anti-establishment when they are not.

read the original abstract

Distrust of public serving institutions and anti-establishment views are on the rise (especially in the U.S.). As people turn to social media for information, it is imperative to understand whether and how social media environments may be contributing to distrust of institutions. In social media, content creators, influencers, and other opinion leaders often position themselves as having expertise and authority on a range of topics from health to politics, and in many cases devalue and dismiss institutional expertise to build a following and increase their own visibility. However, the extent to which this content appears and whether such content increases engagement is unclear. This study analyzes the prevalence of anti-establishment sentiment (AES) on the social media platform TikTok. Despite its popularity as a source of information, TikTok remains relatively understudied and may provide important insights into how people form attitudes towards institutions. We employ a computational approach to label TikTok posts as containing AES or not across topical domains where content creators tend to frame themselves as experts: finance and wellness. As a comparison, we also consider the topic of conspiracy theories, where AES is expected to be common. We find that AES is most prevalent in conspiracy theory content, and relatively rare in content related to the other two topics. However, we find that engagement patterns with such content varies by area, and that there may be platform incentives for users to post content that expresses anti-establishment sentiment.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 1 minor

Summary. The paper examines anti-establishment sentiment (AES) on TikTok in finance, wellness, and conspiracy theory domains. Using a computational labeling method, it reports that AES is most prevalent in conspiracy content and relatively rare in finance and wellness posts, while engagement patterns vary by topic and may reflect platform incentives for AES content.

Significance. If the AES labeling is shown to be accurate and unbiased across domains, the findings would help explain how social media environments contribute to institutional distrust by documenting prevalence and engagement differences, with implications for influencer strategies and platform moderation.

major comments (1)
  1. Methods section: The computational approach for labeling posts as AES or not is described without per-domain human validation, inter-annotator agreement scores, or performance metrics (precision/recall/F1) on held-out samples from finance, wellness, and conspiracy topics. This directly affects the validity of the comparative prevalence claim, as domain-specific bias or misclassification could alter the reported differences.
minor comments (1)
  1. Abstract: The headline findings on prevalence and engagement are stated without any reference to sample sizes, statistical tests, or controls, making it difficult to evaluate the strength of the evidence from the outset.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their constructive feedback on our manuscript. We address the major comment below and will revise the paper to strengthen the methodological validation of our computational labeling approach.

read point-by-point responses
  1. Referee: Methods section: The computational approach for labeling posts as AES or not is described without per-domain human validation, inter-annotator agreement scores, or performance metrics (precision/recall/F1) on held-out samples from finance, wellness, and conspiracy topics. This directly affects the validity of the comparative prevalence claim, as domain-specific bias or misclassification could alter the reported differences.

    Authors: We agree that domain-specific validation is necessary to support the comparative prevalence claims across topics. Our computational labeling method was applied uniformly using the same procedure for consistency, but we acknowledge that this does not substitute for per-domain human evaluation. In the revised manuscript, we will expand the Methods section to include human annotation of held-out samples from each domain (finance, wellness, and conspiracy). We will report inter-annotator agreement scores and performance metrics (precision, recall, and F1) against human judgments for each domain separately. This addition will allow us to quantify and discuss any domain-specific biases or misclassification rates. revision: yes

Circularity Check

0 steps flagged

No significant circularity; empirical labeling and metrics are independent of target claims

full rationale

The paper's central claims rest on applying a computational classifier to TikTok posts across domains, then reporting observed prevalence and engagement differences. No equations, fitted parameters, or self-citations reduce the prevalence finding or engagement patterns to the inputs by construction. The labeling step is presented as an external method applied to data rather than a self-definitional or fitted-input prediction loop. Self-citations, if present, are not load-bearing for the comparative AES results. The derivation chain remains self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Only the abstract is available, so specific free parameters, axioms, or invented entities cannot be extracted. The work implicitly relies on standard assumptions in computational social science regarding the reliability of automated text classification for sentiment detection.

pith-pipeline@v0.9.0 · 5795 in / 1276 out tokens · 65273 ms · 2026-05-21T22:51:57.981872+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.