From Echo Chambers to Epistemic Monoculture: Large Language Models Present Temporally Contingent Partisan Alignments as Knowledge

• URL: http://arxiv.org/abs/2609.07735v1
• Date: Mon, 07 Sep 2026 16:41:56 GMT
• Title: From Echo Chambers to Epistemic Monoculture: Large Language Models Present Temporally Contingent Partisan Alignments as Knowledge
• Abstract summary: We present experiments that exploit a model's training cutoff in 2024.<n>We find that the model presents temporally contingent partisan alignments as knowledge, with no mechanism for distinguishing fact from opinion.
• Score: 0.0
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Large language models (LLMs) are rapidly becoming an interface between citizens and political information. They are often regarded as "a better Google." While this analogy might work for some instances, it is unintuitively problematic for democratic politics. A search engine retrieves human-authored documents, while a language model generates novel text that necessarily embeds invisible framing decisions. Because conveying knowledge involves framing, a system that generates answers cannot serve as a neutral conduit to "all human knowledge." Instead, these systems are becoming a new kind of political intermediary. Mechanistic evidence shows that partisan identity is encoded as a locatable geometric direction inside the Llama 3.1 8B model, and that alignment training masks rather than removes this structure. Building on that evidence, we present steering experiments that exploit a model's training cutoff in 2024. This cutpoint auspiciously falls just before a dramatic realignment in American politics marked by the second Trump administration and the MAHA transformation of health politics, providing us with a natural experiment. We find that the model presents temporally contingent partisan alignments as knowledge, with no mechanism for distinguishing fact from opinion. This reality moves the information environment beyond the echo chamber toward an epistemic monoculture where language models, purporting to summarize "all human knowledge" are, in actuality, simply magnifying the cultural and partisan divides inherent in their training data.

Related papers

• The Amplifying Mirror: Locating and Steering the Partisan Direction inside a Large Language Model [0.0]
  We show that partisan political identity is encoded in the model's activation space.<n>Using 190,491 tweets from sitting members of the U.S. Congress as labeled training data, we train linear probes on the hidden states of the Llama 3.1 8B Instruct model.
  arXiv  Detail & Related papers  (2026-06-07T19:17:02Z)
• The Democratic Ontology Deficit: How AI Systems Fail to Represent What Democracy Requires [0.0]
  This paper identifies and tests the democratic ontology deficit.<n>The model's default ontology is organized under independence rather than civic structure.<n>The deepest deficit is in role: the model's representation of what a person is defaults almost entirely to individual rather than communal identity.
  arXiv  Detail & Related papers  (2026-04-04T21:05:31Z)
• Machine Bullshit: Characterizing the Emergent Disregard for Truth in Large Language Models [57.834711966432685]
  Bullshit, as conceptualized by philosopher Harry Frankfurt, refers to statements made without regard to their truth value.<n>We introduce the Bullshit Index, a novel metric quantifying large language model's indifference to truth.<n>We observe prevalent machine bullshit in political contexts, with weasel words as the dominant strategy.
  arXiv  Detail & Related papers  (2025-07-10T07:11:57Z)
• Generative Exaggeration in LLM Social Agents: Consistency, Bias, and Toxicity [2.3997896447030653]
  We investigate how Large Language Models (LLMs) behave when simulating political discourse on social media.<n>We construct LLM agents based on 1,186 real users, prompting them to reply to politically salient tweets under controlled conditions.<n>We find that richer contextualization improves internal consistency but also amplifies polarization, stylized signals, and harmful language.
  arXiv  Detail & Related papers  (2025-07-01T10:54:51Z)
• Representation Bias in Political Sample Simulations with Large Language Models [54.48283690603358]
  This study seeks to identify and quantify biases in simulating political samples with Large Language Models. Using the GPT-3.5-Turbo model, we leverage data from the American National Election Studies, German Longitudinal Election Study, Zuobiao dataset, and China Family Panel Studies.
  arXiv  Detail & Related papers  (2024-07-16T05:52:26Z)
• DocNet: Semantic Structure in Inductive Bias Detection Models [0.4779196219827508]
  We present DocNet, a novel, inductive, and low-resource document embedding and political bias detection model.<n>We demonstrate that the semantic structure of news articles from opposing political sides, as represented in document-level graph embeddings, have significant similarities.
  arXiv  Detail & Related papers  (2024-06-16T14:51:12Z)
• SADAS: A Dialogue Assistant System Towards Remediating Norm Violations in Bilingual Socio-Cultural Conversations [56.31816995795216]
  Socially-Aware Dialogue Assistant System (SADAS) is designed to ensure that conversations unfold with respect and understanding. Our system's novel architecture includes: (1) identifying the categories of norms present in the dialogue, (2) detecting potential norm violations, (3) evaluating the severity of these violations, and (4) implementing targeted remedies to rectify the breaches.
  arXiv  Detail & Related papers  (2024-01-29T08:54:21Z)
• InfoPattern: Unveiling Information Propagation Patterns in Social Media [59.67008841974645]
  InfoPattern centers on the interplay between language and human ideology. The demo is capable of: (1) red teaming to simulate adversary responses from opposite ideology communities; (2) stance detection to identify the underlying political sentiments in each message; (3) information propagation graph discovery to reveal the evolution of claims across various communities over time.
  arXiv  Detail & Related papers  (2023-11-27T09:12:35Z)
• Getting aligned on representational alignment [93.08284685325674]
  We study the study of representational alignment in cognitive science, neuroscience, and machine learning. Despite their overlapping interests, there is limited knowledge transfer between these fields. We propose a unifying framework that can serve as a common language for research on representational alignment.
  arXiv  Detail & Related papers  (2023-10-18T17:47:58Z)
• Physics of Language Models: Part 3.2, Knowledge Manipulation [51.68385617116854]
  This paper investigates four fundamental knowledge manipulation tasks. We show that language models excel in knowledge retrieval but struggle even in the simplest classification or comparison tasks. Our findings also apply to modern pretrained language models such as GPT-4.
  arXiv  Detail & Related papers  (2023-09-25T17:50:41Z)
• The Face of Populism: Examining Differences in Facial Emotional Expressions of Political Leaders Using Machine Learning [50.24983453990065]
  We use a deep-learning approach to process a sample of 220 YouTube videos of political leaders from 15 different countries.<n>We observe statistically significant differences in the average score of negative emotions between groups of leaders with varying degrees of populist rhetoric.
  arXiv  Detail & Related papers  (2023-04-19T18:32:49Z)
• Large Language Models Can Be Used to Estimate the Latent Positions of Politicians [3.9940425551415597]
  Existing approaches to estimating politicians' latent positions often fail when relevant data is limited. We leverage the embedded knowledge in generative large language models to measure lawmakers' positions along specific political or policy dimensions. We estimate novel measures of U.S. senators' positions on liberal-conservative ideology, gun control, and abortion.
  arXiv  Detail & Related papers  (2023-03-21T17:48:00Z)
• Censorship of Online Encyclopedias: Implications for NLP Models [0.0]
  We show how government repression, censorship, and self-censorship may impact training data and the applications that draw from them. We show that word embeddings trained on Baidu Baike, an online Chinese encyclopedia, have very different associations between adjectives and a range of concepts. Our paper shows how government repression, censorship, and self-censorship may impact training data and the applications that draw from them.
  arXiv  Detail & Related papers  (2021-01-22T19:09:53Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.