Twenty-three researchers argued that the terms of service and enforcement strategies of prominent AI companies discourage good-faith safety evaluation and that researcher access programmes “often do not notify researchers of rejections”; they proposed voluntary legal and technical safe harbours.

SocioLens project,

Claim LD08-SL-C28 (v1.0)

Provenance and status

Claim type: Interpretation
Publication status: Eligible for publication, with qualifiers: interpretation label
Verification state: Checked against the primary source; the source is a preprint. 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: A measured rate of rejections, or that the proposals were adopted
Limitations: Read in arXiv version 1 (March 2024); a later conference version was not read; the paper relays others’ argument that such patchworks create “a veneer of open and responsible research”; it notes that one developer changed its policy after reading a draft
Source (as cited): Longpre, Kapoor, Klyman, Ramaswami, Bommasani et al., “A Safe Harbor for AI Evaluation and Red Teaming,” arXiv:2403.04893, 2024
Source type: Preprint (recorded as “Preprint”)
Institutional position: Independent academic and other researchers (10 of 23 at MIT, Princeton and Stanford)
Maturity of the evidence: Emerging evidence
Recorded date: 16 October 2026
Currentness / correction: This is the current version. No correction or concern is recorded on its sources.

Sources this version cites

Longpre, Kapoor, Klyman, Ramaswami, Bommasani et al., “A Safe Harbor for AI Evaluation and Red Teaming,” arXiv:2403.04893, 2024 (Preprint)
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): INTERPRETATION (as printed: “INTERPRETATION”)
Publication status (coded): ELIGIBLE (INTERPRETATION_LABEL) (as printed: “ELIGIBLE WITH INTERPRETATION LABEL”)
Verification state (coded): CONTENT_VERIFIED_PRIMARY (PREPRINT)
Source type (coded): PREPRINT
Maturity (coded): EMERGING

#	Canonical field	Canonical value (verbatim)	Normalized (Claim Value Vocabulary v1.0.2)
1	CLAIM ID	LD08-SL-C28	—
2	PUBLIC-FACING CLAIM	Twenty-three researchers argued that the terms of service and enforcement strategies of prominent AI companies discourage good-faith safety evaluation and that researcher access programmes “often do not notify researchers of rejections”; they proposed voluntary legal and technical safe harbours.	—
3	CLAIM TYPE	INTERPRETATION	base_types: INTERPRETATION
4	RESEARCH STRAND	B — Data and access	—
5	MATURITY	EMERGING	states: EMERGING
6	SOURCE	Longpre, Kapoor, Klyman, Ramaswami, Bommasani et al., “A Safe Harbor for AI Evaluation and Red Teaming,” arXiv:2403.04893, 2024	source_ids: SRC-SL-LONGPRE-ET-AL-SAFE-HARBOR-ARXIV-2403-04893
7	SOURCE TYPE	Preprint	classes: PREPRINT
8	INSTITUTIONAL POSITION	Independent academic and other researchers (10 of 23 at MIT, Princeton and Stanford)	—
9	SCOPE	Terms of service and access programmes of major AI developers, early 2024	—
10	DESIGN	Policy analysis and proposal	—
11	SAMPLE / DATASET	Not applicable	—
12	MODEL / TOOL / VERSION	Not applicable	—
13	COMPARATOR	Not applicable	—
14	MAIN RESULT	Researcher access depends on developers’ discretion; rejections often go unnotified; proposes legal and technical safe harbours	—
15	INFERENTIAL TARGET	The conditions under which outsiders can test AI systems	—
16	DOES NOT LICENSE	A measured rate of rejections, or that the proposals were adopted	—
17	VERIFICATION STATE	CONTENT VERIFIED PREPRINT	state: CONTENT_VERIFIED_PRIMARY · qualifiers: PREPRINT · verified_by: UNKNOWN
18	LIMITATIONS	Read in arXiv version 1 (March 2024); a later conference version was not read; the paper relays others’ argument that such patchworks create “a veneer of open and responsible research”; it notes that one developer changed its policy after reading a draft	—
19	CONTRADICTORY / SUPPORTING SOURCES	LD08-SL-C29	—
20	GEOGRAPHY / LANGUAGE	US / English	—
21	DISTRIBUTIONAL LEDGER	Not analysed in this claim	—
22	R01 / FIELD-MAP LINKS	LD-08, D3, D6	—
23	WHAT WOULD CHANGE THIS CLAIM?	Adoption of safe harbours, or data on access requests and outcomes	—
24	PUBLICATION STATUS	ELIGIBLE WITH INTERPRETATION LABEL	state: ELIGIBLE · qualifiers: INTERPRETATION_LABEL
25	RECORDED DATE	2026-10-16	—

Field 22 references resolved against Field Map v0.2: LD-08 (LIVING_DEBATE); D3 (DOMAIN); D6 (DOMAIN)

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

Pinned Sources (field 6, version-specific)

“Longpre, Kapoor, Klyman, Ramaswami, Bommasani et al., “A Safe Harbor for AI Evaluation and Red Teaming,” arXiv:2403.04893, 2024” → Longpre et al. (2024), “A Safe Harbor for AI Evaluation…” (SRC-SL-LONGPRE-ET-AL-SAFE-HARBOR-ARXIV-2403-04893--v1.0) — Longpre, Kapoor, Klyman, Ramaswami, Bommasani et al., “A Safe Harbor for AI Evaluation and Red Teaming,” arXiv:2403.04893, 2024; class PREPRINT (explicit public provenance pin)

Cited by Living Debate versions

Who can know about AI? Epistemic dependence and jurisdiction over the field (v0.1) LD-08

Cite this page

SocioLens project. (2026, October 16). Twenty-three researchers argued that the terms of service and enforcement strategies of prominent AI companies discourage good-faith safety evaluation and that researcher access programmes “often do not notify researchers of rejections”; they proposed voluntary legal and technical safe harbours [Claim LD08-SL-C28, version v1.0]. SocioLens. https://sociolens-web.pages.dev/claims/ld08-sl-c28/v1.0/

@misc{sociolens_ld08_sl_c28_v1_0,
  author = {{SocioLens project}},
  title = {Twenty-three researchers argued that the terms of service and enforcement strategies of prominent AI companies discourage good-faith safety evaluation and that researcher access programmes “often do not notify researchers of rejections”; they proposed voluntary legal and technical safe harbours.},
  year = {2026},
  date = {2026-10-16},
  howpublished = {\url{https://sociolens-web.pages.dev/claims/ld08-sl-c28/v1.0/}},
  note = {Claim LD08-SL-C28, 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 9e01b6838f72f0d0abcdc61adae59ea81d9ae949513314626fa00b891969c068.

Repository-authored inputs (not Drive exports)

SocioLens claim — LD08-SL-C28 v1.0 (v1.0) — written in the SocioLens repository by Project Lead (Claude, AI) on 2026-10-16; 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 16 October 2026 to build the LD-08 debate; a new Claim, first version, written from the project's LD-08 source notes, checked against full texts, official web pages or official abstracts. Block “ENTRY 28 — LD08-SL-C28” SHA-256 3a6d7df263d9327251428355804e9336a232ace126a0953a5ea20b8c96a65be7; fixture fixtures/authored/claims/LD08-SL-C28--v1.0.

Builder decisions (not Drive data)

VERSION = v1.0 — basis: Claim version taken from the repository-authored claim fixture version (SocioLens claim — LD08-SL-C28 v1.0); the first version of a new Claim of the LD-08 debate
CURRENT_POINTER = true — basis: repository-authored claim "SocioLens claim — LD08-SL-C28 v1.0" (Project Lead (Claude, AI), 2026-10-16); 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 16 October 2026 to build the LD-08 debate; a new Claim, first version, written from the project's LD-08 source notes, checked against full texts, official web pages or official abstracts
FIELD_6_SOURCE_REFERENCE = SRC-SL-LONGPRE-ET-AL-SAFE-HARBOR-ARXIV-2403-04893--v1.0 — basis: field 6 of ENTRY 28 — LD08-SL-C28 cites "Longpre, Kapoor, Klyman, Ramaswami, Bommasani et al., “A Safe Harbor for AI Evaluation and Red Teaming,” arXiv:2403.04893, 2024"; the only field-6 segment of this ClaimVersion's repository-authored ledger block.
SOURCE_PROVENANCE_PIN = SRC-SL-LONGPRE-ET-AL-SAFE-HARBOR-ARXIV-2403-04893--v1.0 — basis: explicit version-pinned public provenance of this ClaimVersion (Amendment v1.2 §6): the pinned field-6 Source record "Longpre, Kapoor, Klyman, Ramaswami, Bommasani et al., “A Safe Harbor for AI Evaluation and Red Teaming,” arXiv:2403.04893, 2024".