In the customer-support firm where an AI assistant raised productivity most for less-skilled agents, agents who had used it still outperformed their own pre-AI baseline during software outages when it gave no suggestions, more so after longer exposure and closer adherence to its suggestions.

SocioLens project,

Claim LD04-SL-C09 (v1.0)

Research opportunities this claim motivates

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

Does retained skill depend on how routine the task is and who is learning?In routine customer support, agents still beat their pre-AI baseline when the assistant was briefly unavailable: retained gains in a routine task.
Do experienced clinicians lose unaided skill after routine AI use?In customer support, retained gains after AI exposure point the other way, so the direction of the effect on incumbents is not settled across settings.

Provenance and status

Claim type: Estimate
Publication status: Eligible for publication, with qualifiers: version
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: Retention over long periods without the tool; generalisation beyond a routine, scripted task
Limitations: Outages last minutes to hours and are not planned removals; one firm and one task; the published version is Quarterly Journal of Economics 140(2), 2025, and the outage analysis was read in the arXiv v2 version
Source (as cited): Brynjolfsson, Li & Raymond, “Generative AI at Work,” arXiv:2304.11771 v2, November 2024
Source type: Preprint (recorded as “Preprint”)
Institutional position: Independent academic
Maturity of the evidence: Emerging evidence
Recorded date: 14 October 2026
Currentness / correction: This is the current version. No correction or concern is recorded on its sources.

Sources this version cites

Brynjolfsson, Li & Raymond, “Generative AI at Work,” arXiv:2304.11771 v2, November 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): ESTIMATE (as printed: “ESTIMATE”)
Publication status (coded): ELIGIBLE (VERSION) (as printed: “ELIGIBLE WITH VERSION QUALIFIER”)
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	LD04-SL-C09	—
2	PUBLIC-FACING CLAIM	In the customer-support firm where an AI assistant raised productivity most for less-skilled agents, agents who had used it still outperformed their own pre-AI baseline during software outages when it gave no suggestions, more so after longer exposure and closer adherence to its suggestions.	—
3	CLAIM TYPE	ESTIMATE	base_types: ESTIMATE
4	RESEARCH STRAND	B — Skill after the tool is removed	—
5	MATURITY	EMERGING	states: EMERGING
6	SOURCE	Brynjolfsson, Li & Raymond, “Generative AI at Work,” arXiv:2304.11771 v2, November 2024	source_ids: SRC-SL-BRYNJOLFSSON-LI-RAYMOND-ARXIV-2304-11771-V2
7	SOURCE TYPE	Preprint	classes: PREPRINT
8	INSTITUTIONAL POSITION	Independent academic	—
9	SCOPE	One firm’s customer-support agents	—
10	DESIGN	Chat-level comparison during unplanned software outages within a staggered rollout	—
11	SAMPLE / DATASET	5,172 agents	—
12	MODEL / TOOL / VERSION	Generative-AI conversational assistant	—
13	COMPARATOR	The same agents before AI access	—
14	MAIN RESULT	During outages, treated agents beat their pre-AI baseline; the effect grows with exposure and with adherence to AI suggestions; the authors read this as evidence that exposure to AI leads to durable changes in worker skills	—
15	INFERENTIAL TARGET	Whether gains from AI assistance persist when the tool is briefly unavailable	—
16	DOES NOT LICENSE	Retention over long periods without the tool; generalisation beyond a routine, scripted task	—
17	VERIFICATION STATE	CONTENT VERIFIED PREPRINT	state: CONTENT_VERIFIED_PRIMARY · qualifiers: PREPRINT · verified_by: UNKNOWN
18	LIMITATIONS	Outages last minutes to hours and are not planned removals; one firm and one task; the published version is Quarterly Journal of Economics 140(2), 2025, and the outage analysis was read in the arXiv v2 version	—
19	CONTRADICTORY / SUPPORTING SOURCES	Evidence against hollowing in a routine task; contrasts with losses among novices learning new material without guardrails	—
20	GEOGRAPHY / LANGUAGE	Not checked / English	—
21	DISTRIBUTIONAL LEDGER	Retained gains concentrated among agents who followed the AI most	—
22	R01 / FIELD-MAP LINKS	LD-04, LD-02, D2	—
23	WHAT WOULD CHANGE THIS CLAIM?	A planned removal of the tool, or a longer outage, that shows gains fading	—
24	PUBLICATION STATUS	ELIGIBLE WITH VERSION QUALIFIER	state: ELIGIBLE · qualifiers: VERSION
25	RECORDED DATE	2026-10-14	—

Field 22 references resolved against Field Map v0.2: LD-04 (LIVING_DEBATE); LD-02 (LIVING_DEBATE); D2 (DOMAIN)

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

Pinned Sources (field 6, version-specific)

“Brynjolfsson, Li & Raymond, “Generative AI at Work,” arXiv:2304.11771 v2, November 2024” → Brynjolfsson et al. (2024), “Generative AI at Work” (SRC-SL-BRYNJOLFSSON-LI-RAYMOND-ARXIV-2304-11771-V2--v1.0) — Brynjolfsson, Li & Raymond, “Generative AI at Work,” arXiv:2304.11771 v2, November 2024; class PREPRINT (explicit public provenance pin)

Cited by Living Debate versions

Cite this page

SocioLens project. (2026, October 14). In the customer-support firm where an AI assistant raised productivity most for less-skilled agents, agents who had used it still outperformed their own pre-AI baseline during software outages when it gave no suggestions, more so after longer exposure and closer adherence to its suggestions [Claim LD04-SL-C09, version v1.0]. SocioLens. https://sociolens-web.pages.dev/claims/ld04-sl-c09/v1.0/

@misc{sociolens_ld04_sl_c09_v1_0,
  author = {{SocioLens project}},
  title = {In the customer-support firm where an AI assistant raised productivity most for less-skilled agents, agents who had used it still outperformed their own pre-AI baseline during software outages when it gave no suggestions, more so after longer exposure and closer adherence to its suggestions.},
  year = {2026},
  date = {2026-10-14},
  howpublished = {\url{https://sociolens-web.pages.dev/claims/ld04-sl-c09/v1.0/}},
  note = {Claim LD04-SL-C09, 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 df5623ef1073df810adcfcf162323e4b6e20b9114c0d0de577c5bb4f885fa63a.

Repository-authored inputs (not Drive exports)

SocioLens claim — LD04-SL-C09 v1.0 (v1.0) — written in the SocioLens repository by Project Lead (Claude, AI) on 2026-10-14; 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 14 October 2026 to build the LD-04 debate; a new Claim, first version, written from the project's LD-04 source notes, checked against full texts or official abstracts. Block “ENTRY 09 — LD04-SL-C09” SHA-256 1c962e4817a6cb85fbb7fe5c5444446be46cd126b0212f735d95a7ef9a7c1409; fixture fixtures/authored/claims/LD04-SL-C09--v1.0.

Builder decisions (not Drive data)

VERSION = v1.0 — basis: Claim version taken from the repository-authored claim fixture version (SocioLens claim — LD04-SL-C09 v1.0); the first version of a new Claim of the LD-04 debate
CURRENT_POINTER = true — basis: repository-authored claim "SocioLens claim — LD04-SL-C09 v1.0" (Project Lead (Claude, AI), 2026-10-14); 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 14 October 2026 to build the LD-04 debate; a new Claim, first version, written from the project's LD-04 source notes, checked against full texts or official abstracts
FIELD_6_SOURCE_REFERENCE = SRC-SL-BRYNJOLFSSON-LI-RAYMOND-ARXIV-2304-11771-V2--v1.0 — basis: field 6 of ENTRY 09 — LD04-SL-C09 cites "Brynjolfsson, Li & Raymond, “Generative AI at Work,” arXiv:2304.11771 v2, November 2024"; the only field-6 segment of this ClaimVersion's repository-authored ledger block.
SOURCE_PROVENANCE_PIN = SRC-SL-BRYNJOLFSSON-LI-RAYMOND-ARXIV-2304-11771-V2--v1.0 — basis: explicit version-pinned public provenance of this ClaimVersion (Amendment v1.2 §6): the pinned field-6 Source record "Brynjolfsson, Li & Raymond, “Generative AI at Work,” arXiv:2304.11771 v2, November 2024".