Across 27 language models, 155 topics and 12 countries, the epistemic diversity of models’ answers rose substantially over three years, but every model was less diverse than a web-search baseline, larger models were less diverse than smaller ones, and for country-specific topics models’ built-in knowledge reflected English more than local-language knowledge.

SocioLens project,

Claim LD09-SL-C32 (v1.0)

Provenance and status

Claim type: Estimate
Publication status: Eligible for publication, with a qualifier
Verification state: Checked against the abstract of the primary source only. Who checked this claim is not recorded for it; claim checks so far were AI-led, and no person has checked claims against their sources.
Does not license: Effects on what people read or believe
Limitations: Official abstract only (version 7); the abstract’s wording changed across versions
Source (as cited): Wright, Masud, Moore et al., “What and whose knowledge? Measuring epistemic diversity in large language models,” arXiv:2510.04226, accepted to EMNLP 2026
Source type: Peer-reviewed conference paper (recorded as “Peer-reviewed conference paper (accepted)”)
Institutional position: Independent academic (University of Copenhagen-led)
Maturity of the evidence: Emerging evidence
Recorded date: 15 October 2026
Currentness / correction: This is the current version. No correction or concern is recorded on its sources.

Sources this version cites

Wright, Masud, Moore et al., “What and whose knowledge? Measuring epistemic diversity in large language models,” arXiv:2510.04226, accepted to EMNLP 2026 (Peer-reviewed conference paper)
Read the original

Version

This page shows version v1.0. Its record never changes, even when newer evidence arrives; the site may add a note to the page or withhold a figure not cleared for public use.

This pinned version is also the current version.

Versions: v1.0 · current

Raw record: every recorded field, with its coded values

All 25 canonical Claim fields, verbatim, with the coded values derived from them. The plain-language labels above are read from these.

Claim type (coded): ESTIMATE (as printed: “ESTIMATE”)
Publication status (coded): ELIGIBLE (QUALIFIED) (as printed: “ELIGIBLE WITH QUALIFIER”)
Verification state (coded): CONTENT_VERIFIED_PRIMARY (ABSTRACT_ONLY)
Source type (coded): PEER_REVIEWED_CONFERENCE_PAPER
Maturity (coded): EMERGING

#	Canonical field	Canonical value (verbatim)	Normalized (Claim Value Vocabulary v1.0.2)
1	CLAIM ID	LD09-SL-C32	—
2	PUBLIC-FACING CLAIM	Across 27 language models, 155 topics and 12 countries, the epistemic diversity of models’ answers rose substantially over three years, but every model was less diverse than a web-search baseline, larger models were less diverse than smaller ones, and for country-specific topics models’ built-in knowledge reflected English more than local-language knowledge.	—
3	CLAIM TYPE	ESTIMATE	base_types: ESTIMATE
4	RESEARCH STRAND	C — Systems, decisions and shared knowledge	—
5	MATURITY	EMERGING	states: EMERGING
6	SOURCE	Wright, Masud, Moore et al., “What and whose knowledge? Measuring epistemic diversity in large language models,” arXiv:2510.04226, accepted to EMNLP 2026	source_ids: SRC-SL-WRIGHT-ET-AL-EPISTEMIC-DIVERSITY-ARXIV-2510-04226
7	SOURCE TYPE	Peer-reviewed conference paper (accepted)	classes: PEER_REVIEWED_CONFERENCE_PAPER
8	INSTITUTIONAL POSITION	Independent academic (University of Copenhagen-led)	—
9	SCOPE	Answers of language models on general topics in 12 countries	—
10	DESIGN	Measurement of the diversity of claims in model answers against a search baseline	—
11	SAMPLE / DATASET	1.7 million responses and 70 million claims	—
12	MODEL / TOOL / VERSION	27 language models	—
13	COMPARATOR	A web-search baseline	—
14	MAIN RESULT	Diversity up over three years; all models below the search baseline; retrieval improves diversity; larger models less diverse; built-in knowledge skewed to English	—
15	INFERENTIAL TARGET	Whether answer engines narrow the range of knowledge people meet	—
16	DOES NOT LICENSE	Effects on what people read or believe	—
17	VERIFICATION STATE	CONTENT VERIFIED ABSTRACT	state: CONTENT_VERIFIED_PRIMARY · qualifiers: ABSTRACT_ONLY · verified_by: UNKNOWN
18	LIMITATIONS	Official abstract only (version 7); the abstract’s wording changed across versions	—
19	CONTRADICTORY / SUPPORTING SOURCES	Not applicable	—
20	GEOGRAPHY / LANGUAGE	Multi-country	—
21	DISTRIBUTIONAL LEDGER	Local-language knowledge is under-represented relative to English	—
22	R01 / FIELD-MAP LINKS	LD-09, D3	—
23	WHAT WOULD CHANGE THIS CLAIM?	Studies of what users actually read when they use answer engines	—
24	PUBLICATION STATUS	ELIGIBLE WITH QUALIFIER	state: ELIGIBLE · qualifiers: QUALIFIED
25	RECORDED DATE	2026-10-15	—

Field 22 references resolved against Field Map v0.2: LD-09 (LIVING_DEBATE); D3 (DOMAIN)

The sources as recorded, with their record IDs and coded source types:

Pinned Sources (field 6, version-specific)

“Wright, Masud, Moore et al., “What and whose knowledge? Measuring epistemic diversity in large language models,” arXiv:2510.04226, accepted to EMNLP 2026” → Wright et al. (2026), “What and whose knowledge?…” (SRC-SL-WRIGHT-ET-AL-EPISTEMIC-DIVERSITY-ARXIV-2510-04226--v1.0) — Wright, Masud, Moore et al., “What and whose knowledge? Measuring epistemic diversity in large language models,” arXiv:2510.04226, accepted to EMNLP 2026; class PEER_REVIEWED_CONFERENCE_PAPER (explicit public provenance pin)

Cited by Living Debate versions

Individual gain, collective narrowing? Does AI make science, creative work and shared knowledge more alike? (v0.1) LD-09

Cite this page

SocioLens project. (2026, October 15). Across 27 language models, 155 topics and 12 countries, the epistemic diversity of models’ answers rose substantially over three years, but every model was less diverse than a web-search baseline, larger models were less diverse than smaller ones, and for country-specific topics models’ built-in knowledge reflected English more than local-language knowledge [Claim LD09-SL-C32, version v1.0]. SocioLens. https://sociolens-web.pages.dev/claims/ld09-sl-c32/v1.0/

@misc{sociolens_ld09_sl_c32_v1_0,
  author = {{SocioLens project}},
  title = {Across 27 language models, 155 topics and 12 countries, the epistemic diversity of models’ answers rose substantially over three years, but every model was less diverse than a web-search baseline, larger models were less diverse than smaller ones, and for country-specific topics models’ built-in knowledge reflected English more than local-language knowledge.},
  year = {2026},
  date = {2026-10-15},
  howpublished = {\url{https://sociolens-web.pages.dev/claims/ld09-sl-c32/v1.0/}},
  note = {Claim LD09-SL-C32, version v1.0}
}

The link points to this version, whose record never changes. For a finding, cite the original study as well: SocioLens summarizes research and does not replace it.

Research-of-record provenance

Materialized by sociolens-materializer 0.6.0 at 2026-09-28T20:00:00.000Z; normalization contract SocioLens — Claim Value Vocabulary v1.0.2 — Pre-Spike Materialization Contract (v1.0.2). Content seal SHA-256 c79721c678719e5ac6ca81023b5219ce2e26da1dedddc0a9952f3124ee8ee7a5.

Repository-authored inputs (not Drive exports)

SocioLens claim — LD09-SL-C32 v1.0 (v1.0) — written in the SocioLens repository by Project Lead (Claude, AI) on 2026-10-15; no Drive identity exists. Authority: AGENTS.md "Authority" (the Project Lead decides research content within the research-integrity rules) and the Owner's choice in the SocioLens project chat on 15 October 2026 to build the LD-09 debate; a new Claim, first version, written from the project's LD-09 source notes, checked against full texts or official abstracts. Block “ENTRY 32 — LD09-SL-C32” SHA-256 53455ac5cff58b8028778667c7175ed585216892e3efdf15eb5e88d20751b559; fixture fixtures/authored/claims/LD09-SL-C32--v1.0.

Builder decisions (not Drive data)

VERSION = v1.0 — basis: Claim version taken from the repository-authored claim fixture version (SocioLens claim — LD09-SL-C32 v1.0); the first version of a new Claim of the LD-09 debate
CURRENT_POINTER = true — basis: repository-authored claim "SocioLens claim — LD09-SL-C32 v1.0" (Project Lead (Claude, AI), 2026-10-15); basis: AGENTS.md "Authority" (the Project Lead decides research content within the research-integrity rules) and the Owner's choice in the SocioLens project chat on 15 October 2026 to build the LD-09 debate; a new Claim, first version, written from the project's LD-09 source notes, checked against full texts or official abstracts
FIELD_6_SOURCE_REFERENCE = SRC-SL-WRIGHT-ET-AL-EPISTEMIC-DIVERSITY-ARXIV-2510-04226--v1.0 — basis: field 6 of ENTRY 32 — LD09-SL-C32 cites "Wright, Masud, Moore et al., “What and whose knowledge? Measuring epistemic diversity in large language models,” arXiv:2510.04226, accepted to EMNLP 2026"; the only field-6 segment of this ClaimVersion's repository-authored ledger block.
SOURCE_PROVENANCE_PIN = SRC-SL-WRIGHT-ET-AL-EPISTEMIC-DIVERSITY-ARXIV-2510-04226--v1.0 — basis: explicit version-pinned public provenance of this ClaimVersion (Amendment v1.2 §6): the pinned field-6 Source record "Wright, Masud, Moore et al., “What and whose knowledge? Measuring epistemic diversity in large language models,” arXiv:2510.04226, accepted to EMNLP 2026".