Claim LD04-SL-C12 (v1.0)

In a randomized experiment with 52 Python programmers learning an unfamiliar library, those given an AI assistant scored 17% lower on a later quiz of concepts, code reading and debugging, with no significant speed gain; participants who delegated code to the AI scored lowest.

Research opportunities this claim motivates

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

Do professionals trained with AI lose the skill without it a year later?Programmers learning a new library with AI scored 17% lower on a quiz straight afterwards, and those who delegated code scored lowest; one session only.
Does retained skill depend on how routine the task is and who is learning?Novices learning a new programming library with AI scored lower without it, and delegation predicted the lowest scores.
What AI set-ups do employers actually give trainees and junior staff?Learners who used AI for concepts scored well and those who delegated scored lowest, so rules about how juniors use AI could matter as much as access.

Provenance and status

Claim type: ESTIMATE (as printed: “ESTIMATE”)
Publication status: ELIGIBLE (PREPRINT) (as printed: “ELIGIBLE WITH PREPRINT QUALIFIER”)
Verification state: CONTENT_VERIFIED_PRIMARY (PREPRINT) (as printed: “CONTENT VERIFIED PREPRINT”; verified by: UNKNOWN)
Does not license: Long-term skill loss; effects on experienced workers in their own domain
Source (field 6, as printed): Shen & Tamkin, “How AI Impacts Skill Formation,” arXiv:2601.20245 v2, February 2026
Source type (field 7): PREPRINT (as printed: “Preprint”)
Institutional position: AI developer-affiliated research programme
Maturity (claim evidence / publication maturity): EMERGING (as printed: “EMERGING”)
Recorded date: 2026-10-14
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--LD04-SL-C12.

Versions:v1.0 · current

Pinned Sources (field 6, version-specific)

“Shen & Tamkin, “How AI Impacts Skill Formation,” arXiv:2601.20245 v2, February 2026” → SRC-SL-SHEN-TAMKIN-ARXIV-2601-20245--v1.0 — Shen & Tamkin, “How AI Impacts Skill Formation,” arXiv:2601.20245 v2, February 2026; class PREPRINT (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	LD04-SL-C12	—
2	PUBLIC-FACING CLAIM	In a randomized experiment with 52 Python programmers learning an unfamiliar library, those given an AI assistant scored 17% lower on a later quiz of concepts, code reading and debugging, with no significant speed gain; participants who delegated code to the AI scored lowest.	—
3	CLAIM TYPE	ESTIMATE	base_types: ESTIMATE
4	RESEARCH STRAND	B — Skill after the tool is removed	—
5	MATURITY	EMERGING	states: EMERGING
6	SOURCE	Shen & Tamkin, “How AI Impacts Skill Formation,” arXiv:2601.20245 v2, February 2026	source_ids: SRC-SL-SHEN-TAMKIN-ARXIV-2601-20245
7	SOURCE TYPE	Preprint	classes: PREPRINT
8	INSTITUTIONAL POSITION	AI developer-affiliated research programme	—
9	SCOPE	Crowd-recruited programmers with over a year of Python and no prior use of the Trio library; one session	—
10	DESIGN	Randomized experiment (26 per arm) followed by a 27-point quiz without AI	—
11	SAMPLE / DATASET	52 participants	—
12	MODEL / TOOL / VERSION	AI chat assistant	—
13	COMPARATOR	Participants learning without AI	—
14	MAIN RESULT	AI group 4.15 points (17%) lower on the quiz (Cohen’s d = 0.738, p = 0.010); largest gap on debugging; no significant speed gain; high scorers used AI for concepts, low scorers delegated	—
15	INFERENTIAL TARGET	Immediate learning of a new programming skill with and without AI help	—
16	DOES NOT LICENSE	Long-term skill loss; effects on experienced workers in their own domain	—
17	VERIFICATION STATE	CONTENT VERIFIED PREPRINT	state: CONTENT_VERIFIED_PRIMARY · qualifiers: PREPRINT · verified_by: UNKNOWN
18	LIMITATIONS	Small sample; one library and one session; the quiz has no code-writing items; conflict of interest: the authors worked through the Anthropic Fellows programme	—
19	CONTRADICTORY / SUPPORTING SOURCES	Consistent with school-mathematics losses without guardrails and with the substitution–complement split in classroom experiments	—
20	GEOGRAPHY / LANGUAGE	Not checked / English	—
21	DISTRIBUTIONAL LEDGER	How AI is used (concepts vs delegation) separates learners more than access does	—
22	R01 / FIELD-MAP LINKS	LD-04, D2	—
23	WHAT WOULD CHANGE THIS CLAIM?	Larger replications, or delayed tests that show no lasting difference	—
24	PUBLICATION STATUS	ELIGIBLE WITH PREPRINT QUALIFIER	state: ELIGIBLE · qualifiers: PREPRINT
25	RECORDED DATE	2026-10-14	—

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

Cited by Living Debate versions

LD-04 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 7c898117811129320a61947d5a4eb0cc9572a89e6bce28a433c08af2b31ab816.

Repository-authored inputs (not Drive exports)

SocioLens claim — LD04-SL-C12 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 12 — LD04-SL-C12” SHA-256 7b03c212aadc5d3eb8e30e2da435aa64f3e2a883457e0f5992c0daa84a85c152; fixture fixtures/authored/claims/LD04-SL-C12--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-C12 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-C12 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-SHEN-TAMKIN-ARXIV-2601-20245--v1.0 — basis: field 6 of ENTRY 12 — LD04-SL-C12 cites "Shen & Tamkin, “How AI Impacts Skill Formation,” arXiv:2601.20245 v2, February 2026"; the only field-6 segment of this ClaimVersion's repository-authored ledger block.
SOURCE_PROVENANCE_PIN = SRC-SL-SHEN-TAMKIN-ARXIV-2601-20245--v1.0 — basis: explicit version-pinned public provenance of this ClaimVersion (Amendment v1.2 §6): the pinned field-6 Source record "Shen & Tamkin, “How AI Impacts Skill Formation,” arXiv:2601.20245 v2, February 2026".