Model identity is time-indexed: proprietary services can change materially across deployed snapshots, so exact version and date are part of the measurement protocol.

SocioLens project,

Claim LD07-HQ-C12 (v1.0)

Model identity is time-indexed: proprietary services can change materially across deployed snapshots, so exact version and date are part of the measurement protocol.

Research opportunities this claim motivates

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

Do pinned model snapshots keep social-science conclusions stable over time?Model identity is time-indexed; exact version and date belong in the protocol.
Does quantization or the serving engine change social-science findings?The instrument is defined by more than its name; deployment is part of its identity.
Does asking in Greek change how AI "silicon samples" answer as Greek respondents?Proprietary models change across deployed snapshots, so the model version and date are part of the measurement.

Provenance and status

Claim type: Fact
Publication status: Eligible for publication, with qualifiers: historical snapshot
Verification state: Checked against the primary source. 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: Extrapolation of same drift magnitude to all current pinned models; claim open weights solve everything
Limitations: 2023 snapshots; provider-specific historical example
Source (as cited): Chen, Zaharia & Zou, “How is ChatGPT’s behavior changing over time?”, Harvard Data Science Review 6(2), 2024
Source type: Peer-reviewed journal article (recorded as “Peer-reviewed journal article”)
Institutional position: Academic
Maturity of the evidence: Emerging evidence (foundational)
Recorded date: 25 September 2026
Currentness / correction: This is the current version. No correction or concern is recorded on its sources.

Sources this version cites

Chen, Zaharia & Zou, “How is ChatGPT’s behavior changing over time?,” Harvard Data Science Review 6(2) (2024) (Peer-reviewed journal article)
Read the original

Version

This is the current version, v1.0. Earlier versions stay available, and their records never change.

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): FACT (as printed: “FACT”)
Publication status (coded): ELIGIBLE (HISTORICAL_SNAPSHOT) (as printed: “ELIGIBLE WITH HISTORICAL-SNAPSHOT QUALIFIER”)
Verification state (coded): CONTENT_VERIFIED_PRIMARY
Source type (coded): PEER_REVIEWED_JOURNAL_ARTICLE
Maturity (coded): EMERGING (FOUNDATIONAL)

#	Canonical field	Canonical value (verbatim)	Normalized (Claim Value Vocabulary v1.0.2)
1	CLAIM ID	LD07-HQ-C12	—
2	PUBLIC-FACING CLAIM	Model identity is time-indexed: proprietary services can change materially across deployed snapshots, so exact version and date are part of the measurement protocol.	—
3	CLAIM TYPE	FACT	base_types: FACT
4	RESEARCH STRAND	F	—
5	MATURITY	EMERGING / FOUNDATIONAL	states: EMERGING · qualifiers: FOUNDATIONAL
6	SOURCE	Chen, Zaharia & Zou, “How is ChatGPT’s behavior changing over time?”, Harvard Data Science Review 6(2), 2024	source_ids: SRC-LD07-CHEN-ZAHARIA-ZOU-DRIFT-2024
7	SOURCE TYPE	Peer-reviewed journal article	classes: PEER_REVIEWED_JOURNAL_ARTICLE
8	INSTITUTIONAL POSITION	Academic	—
9	SCOPE	March vs June 2023 GPT-3.5/GPT-4 snapshots	—
10	DESIGN	Longitudinal service audit across seven task families	—
11	SAMPLE / DATASET	Math, safety, opinion surveys, knowledge, code, USMLE, visual reasoning	—
12	MODEL / TOOL / VERSION	GPT-3.5 and GPT-4 March/June 2023	—
13	COMPARATOR	Same named services at different dates	—
14	MAIN RESULT	Material behavior/performance shifts; final HDSR example GPT-4 prime/composite accuracy 84%→51%; willingness on opinion/sensitive questions changed	—
15	INFERENTIAL TARGET	Reproducibility/non-stationarity	—
16	DOES NOT LICENSE	Extrapolation of same drift magnitude to all current pinned models; claim open weights solve everything	—
17	VERIFICATION STATE	CLEARED	state: CONTENT_VERIFIED_PRIMARY · verified_by: UNKNOWN
18	LIMITATIONS	2023 snapshots; provider-specific historical example	—
19	CONTRADICTORY / SUPPORTING SOURCES	Later reproducibility literature to update magnitude	—
20	GEOGRAPHY / LANGUAGE	English task set	—
21	DISTRIBUTIONAL LEDGER	Not assessed	—
22	R01 / FIELD-MAP LINKS	T1, T7, R-1, D9	—
23	WHAT WOULD CHANGE THIS CLAIM?	Long-term proof that exact proprietary snapshots remain behaviorally stable and archivally accessible	—
24	PUBLICATION STATUS	ELIGIBLE WITH HISTORICAL-SNAPSHOT QUALIFIER	state: ELIGIBLE · qualifiers: HISTORICAL_SNAPSHOT
25	RECORDED DATE	2026-09-25	—

Field 22 references resolved against Field Map v0.2: T1 (TENSION); T7 (TENSION); R-1 (LENS); D9 (DOMAIN)

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

Pinned Sources (field 6, version-specific)

“Chen, Zaharia & Zou, “How is ChatGPT’s behavior changing over time?”, Harvard Data Science Review 6(2), 2024” → Chen et al. (2024), “How is ChatGPT’s behavior changing…” (SRC-LD07-CHEN-ZAHARIA-ZOU-DRIFT-2024--v1.0) — Chen, Zaharia & Zou, “How is ChatGPT’s behavior changing over time?,” Harvard Data Science Review 6(2) (2024); class PEER_REVIEWED_JOURNAL_ARTICLE (explicit public provenance pin)

Cited by Living Debate versions

AI in the research apparatus (v0.1) LD-07
AI in the research apparatus (v0.2) LD-07

Cite this page

SocioLens project. (2026, September 25). Model identity is time-indexed: proprietary services can change materially across deployed snapshots, so exact version and date are part of the measurement protocol [Claim LD07-HQ-C12, version v1.0]. SocioLens. https://sociolens-web.pages.dev/claims/ld07-hq-c12/v1.0/

@misc{sociolens_ld07_hq_c12_v1_0,
  author = {{SocioLens project}},
  title = {Model identity is time-indexed: proprietary services can change materially across deployed snapshots, so exact version and date are part of the measurement protocol.},
  year = {2026},
  date = {2026-09-25},
  howpublished = {\url{https://sociolens-web.pages.dev/claims/ld07-hq-c12/v1.0/}},
  note = {Claim LD07-HQ-C12, 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 e9668a837581a16b55aa08d887345c7997aabbe48ece57d6e036cace0cd07d74.

Frozen Drive-derived inputs

R03 — LD-07 Canonical Claim Ledger v1.0 — LOCKED (v1.0) — Drive ID 1xVz_LrQAA8EA1tPhVWIrDgOUU3Zidlg9; revision: NOT_EXPOSED_BY_DRIVE_TEXT_EXPORT; Drive modified 2026-09-29T04:31:14.120Z; block “ENTRY 12 — LD07-HQ-C12” SHA-256 1fc84c068224a516cbd25ebe1523cf72a4a0e56c55866755181ee58e50e3fc1d; SocioLens — Programme Status & Authority Register v0.6 status “CURRENT.”; frozen fixture fixtures/frozen/r03-ld07-claim-ledger-v1.0-canonical.

Builder decisions (not Drive data)

VERSION = v1.0 — basis: Claim version taken from the canonical ledger artifact version (R03 — LD-07 Canonical Claim Ledger v1.0); Implementation Architecture Selection §5
CURRENT_POINTER = true — basis: SocioLens — Programme Status & Authority Register v0.6 status for "R03 — LD-07 Canonical Claim Ledger v1.0 — LOCKED": "CURRENT."
FIELD_6_SOURCE_REFERENCE = SRC-LD07-CHEN-ZAHARIA-ZOU-DRIFT-2024--v1.0 — basis: field 6 of ENTRY 12 — LD07-HQ-C12 cites "Chen, Zaharia & Zou, “How is ChatGPT’s behavior changing over time?”, Harvard Data Science Review 6(2), 2024"; the frozen field-6 segment of this ClaimVersion's ledger block; Drive-derived identity authority (Issue #39 Source inventory): Batch 04 §2; ledger C12 fields 6/7/8; KEY EVIDENCE.
SOURCE_PROVENANCE_PIN = SRC-LD07-CHEN-ZAHARIA-ZOU-DRIFT-2024--v1.0 — basis: Issue #5 proof rendering (Architecture Proof Protocol v0.1 AP-08 / Amendment v1.2 §6): explicit version-pinned public provenance of this ClaimVersion — the pinned field-6 Source record "Chen, Zaharia & Zou, “How is ChatGPT’s behavior changing over time?”, Harvard Data Science Review 6(2), 2024"; Drive-derived identity authority (Issue #39 Source inventory): Batch 04 §2; ledger C12 fields 6/7/8; KEY EVIDENCE.