---
id: "claim-human-bottleneck"
type: "claim"
source_timestamps: ["§ Engage Your People"]
tags: ["sabotage", "roi", "workforce"]
related: ["concept-ai-sabotage", "contrarian-employee-sabotage", "action-appoint-ai-champions", "action-empower-citizen-developers"]
confidence: "high"
testable: true
speakers: ["Cyril Bouquet", "Christopher J. Wright", "Julian Nolan"]
enrichment_status: "General claim (human resistance > algorithmic weakness) strongly supported; the 31%/10% figures and named case metrics are unverified/source-specific."
source_url: "https://hbr.org/2026/01/match-your-ai-strategy-to-your-organizations-reality"
source_title: "Match Your AI Strategy to Your Organization's Reality"
sources: ["spine"]
sourceVaultSlug: "hbr-seg-spine"
originDay: 1
articleStem: "hbr-sig-55-match-ai-strategy-to-reality"
sourceUrl: "https://hbr.org/2026/01/match-your-ai-strategy-to-your-organizations-reality"
sourceTitle: "Match Your AI Strategy to Your Organization’s Reality"
---
# Employee resistance actively sabotages AI ROI

**Claim (confidence: high, testable).** Human resistance is a more significant barrier to AI success than algorithmic weakness. Citing a survey of 1,600 enterprise users, the authors report **31%** of employees actively push back on AI initiatives and **10%** intentionally sabotage them (tampering with metrics or generating low-quality outputs). This resistance directly translates to delayed implementations and lost savings, making change management a hard operational prerequisite for AI ROI. See [[concept-ai-sabotage]].

**Enrichment caveat.** The *general* claim is strongly supported by AI-adoption and change-management literature (fear of job loss → resistance, under-use, shadow processes). The *specific sabotage percentages* and the named Writer survey are not independently corroborated; most organizational research frames resistance as passive non-use or work-arounds rather than explicit malicious tampering. Accept the directional insight; treat the 10% prevalence and named case metrics ([[org-rent-a-mac]], [[org-colgate-palmolive]]) as unverified, source-specific data. Contrarian framing: [[contrarian-employee-sabotage]].


## Related across articles
- [[claim-forced-adoption-workslop]]
- [[claim-people-process-value]]
- [[concept-ai-sabotage]]