Claim LD09-SL-C05 (v1.0)

Across arXiv, bioRxiv and SSRN preprints (2018–2024), authors whose abstracts first showed signs of language-model writing then posted 36.2%, 52.9% and 59.8% more papers respectively; for their manuscripts, more complex writing went with lower odds of publication, the reverse of the pattern for human-written ones; and adopters drew on “more diverse prior work”, being 11.9% more likely to cite books and citing younger documents.

Research opportunities this claim motivates

What this claim leaves open for research. Each is a hypothesis about a gap, not a finding.

Does language-model use narrow what whole fields study, including the social sciences?Language-model adopters posted more preprints, were more likely to cite books and cited younger documents.

Provenance and status

Claim type: ESTIMATE (as printed: “ESTIMATE”)
Publication status: ELIGIBLE (SOURCE_POSITION) (as printed: “ELIGIBLE WITH SOURCE-POSITION QUALIFIER”)
Verification state: CONTENT_VERIFIED_PRIMARY (as printed: “CONTENT VERIFIED”; verified by: UNKNOWN)
Does not license: A causal effect (the authors “do not provide causal identification”), any change in research quality, or a collective narrowing
Source (field 6, as printed): Kusumegi, Yang, Ginsparg, de Vaan, Stuart & Yin, “Scientific production in the era of large language models,” Science 390, 2025 (author version arXiv:2601.13187)
Source type (field 7): PEER_REVIEWED_JOURNAL_ARTICLE (as printed: “Peer-reviewed journal article (author version)”)
Institutional position: Independent academic (Cornell University and UC Berkeley Haas School of Business); a co-author founded arXiv, whose access logs the study uses
Maturity (claim evidence / publication maturity): CONTESTED (as printed: “CONTESTED”)
Recorded date: 2026-10-15
Currentness / correction: This is the current version. No correction or concern relation is recorded on its Sources.

Version

Current version v1.0, resolved through the explicit current pointer CLAIM--LD09-SL-C05.

Versions:v1.0 · current

Pinned Sources (field 6, version-specific)

“Kusumegi, Yang, Ginsparg, de Vaan, Stuart & Yin, “Scientific production in the era of large language models,” Science 390, 2025 (author version arXiv:2601.13187)” → SRC-SL-KUSUMEGI-ET-AL-SCIENCE-2025--v1.0 — Kusumegi, Yang, Ginsparg, de Vaan, Stuart & Yin, “Scientific production in the era of large language models,” Science 390, 2025 (author version arXiv:2601.13187); class PEER_REVIEWED_JOURNAL_ARTICLE (explicit public provenance pin)

All 25 canonical Claim fields (raw values and normalized values)

#	Canonical field	Canonical value (verbatim)	Normalized (Claim Value Vocabulary v1.0.2)
1	CLAIM ID	LD09-SL-C05	—
2	PUBLIC-FACING CLAIM	Across arXiv, bioRxiv and SSRN preprints (2018–2024), authors whose abstracts first showed signs of language-model writing then posted 36.2%, 52.9% and 59.8% more papers respectively; for their manuscripts, more complex writing went with lower odds of publication, the reverse of the pattern for human-written ones; and adopters drew on “more diverse prior work”, being 11.9% more likely to cite books and citing younger documents.	—
3	CLAIM TYPE	ESTIMATE	base_types: ESTIMATE
4	RESEARCH STRAND	A — Science and scholarship	—
5	MATURITY	CONTESTED	states: CONTESTED
6	SOURCE	Kusumegi, Yang, Ginsparg, de Vaan, Stuart & Yin, “Scientific production in the era of large language models,” Science 390, 2025 (author version arXiv:2601.13187)	source_ids: SRC-SL-KUSUMEGI-ET-AL-SCIENCE-2025
7	SOURCE TYPE	Peer-reviewed journal article (author version)	classes: PEER_REVIEWED_JOURNAL_ARTICLE
8	INSTITUTIONAL POSITION	Independent academic (Cornell University and UC Berkeley Haas School of Business); a co-author founded arXiv, whose access logs the study uses	—
9	SCOPE	Preprints in the natural sciences, computer science, economics and, through SSRN, the social sciences, law and humanities	—
10	DESIGN	Stacked event study around each author’s first flagged month; difference-in-differences for citation behaviour	—
11	SAMPLE / DATASET	About 2.1 million preprints; author panels of 109,965 (arXiv), 43,218 (bioRxiv) and 15,370 (SSRN); 246 million arXiv access logs	—
12	MODEL / TOOL / VERSION	The authors’ own detector of language-model text in abstracts	—
13	COMPARATOR	The same authors before their first flagged paper, and similar authors not flagged as users	—
14	MAIN RESULT	Output +36.2% (arXiv), +52.9% (bioRxiv), +59.8% (SSRN), ranging 23.7–89.3% by field and author background; for language-model-assisted manuscripts, more complex writing predicts lower odds of publication, reversing the pattern for human-written ones; adopters are 11.9% more likely to cite books (not significant on SSRN), cite documents 0.379 years younger on average, with no evidence that the cited work is more impactful	—
15	INFERENTIAL TARGET	How language-model use changes the quantity of scientific output and the reading behind it	—
16	DOES NOT LICENSE	A causal effect (the authors “do not provide causal identification”), any change in research quality, or a collective narrowing	—
17	VERIFICATION STATE	CONTENT VERIFIED	state: CONTENT_VERIFIED_PRIMARY · verified_by: UNKNOWN
18	LIMITATIONS	Read in the authors’ version on arXiv, not the typeset journal text; the detector misses heavily edited text; the dating of adoption is disputed (LD09-SL-C06); arXiv’s founder is a co-author and arXiv’s own logs are used	—
19	CONTRADICTORY / SUPPORTING SOURCES	Method dispute: LD09-SL-C06; earlier narrowing with online search: LD09-SL-C03	—
20	GEOGRAPHY / LANGUAGE	Global / English	—
21	DISTRIBUTIONAL LEDGER	The largest gains were for authors with Asian names at institutions in Asia (up to 89.3%), the smallest for authors with Caucasian names in English-speaking countries (23.7%)	—
22	R01 / FIELD-MAP LINKS	LD-09, D3	—
23	WHAT WOULD CHANGE THIS CLAIM?	A design that dates adoption independently of output, or evidence that the extra papers are not published	—
24	PUBLICATION STATUS	ELIGIBLE WITH SOURCE-POSITION QUALIFIER	state: ELIGIBLE · qualifiers: SOURCE_POSITION
25	RECORDED DATE	2026-10-15	—

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

Cited by Living Debate versions

LD-09 v0.1

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 4305dfb9cd86b7c1e4c50d5ddac7399e45476e534db6ff582af6ec023c3eeca0.

Repository-authored inputs (not Drive exports)

SocioLens claim — LD09-SL-C05 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 05 — LD09-SL-C05” SHA-256 6e38aaff612e96e729224f56a95cb310f2725c57b4a1651b08487c67e21bc59a; fixture fixtures/authored/claims/LD09-SL-C05--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-C05 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-C05 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-KUSUMEGI-ET-AL-SCIENCE-2025--v1.0 — basis: field 6 of ENTRY 05 — LD09-SL-C05 cites "Kusumegi, Yang, Ginsparg, de Vaan, Stuart & Yin, “Scientific production in the era of large language models,” Science 390, 2025 (author version arXiv:2601.13187)"; the only field-6 segment of this ClaimVersion's repository-authored ledger block.
SOURCE_PROVENANCE_PIN = SRC-SL-KUSUMEGI-ET-AL-SCIENCE-2025--v1.0 — basis: explicit version-pinned public provenance of this ClaimVersion (Amendment v1.2 §6): the pinned field-6 Source record "Kusumegi, Yang, Ginsparg, de Vaan, Stuart & Yin, “Scientific production in the era of large language models,” Science 390, 2025 (author version arXiv:2601.13187)".