---
type: "synthesis"
tags: ["adoption-gap", "leadership", "cross-cutting-diagnosis"]
articles: ["a040", "a042", "a052", "a078"]
id: "cross-adoption-perception-gap"
sources: ["adoption"]
sourceVaultSlug: "hbr-seg-adoption"
originDay: 9
articleStem: "hbr-seg-adoption"
sourceUrl: "(unified vault: 11 sources)"
sourceTitle: "HBR — People Ⅲ-A · Adoption / trust / literacy / psych-safety"
---
Across the corpus, the single most-repeated *empirical* finding is a chasm between how leaders and workers experience the same AI rollout. It is the shared diagnosis nearly every prescription then treats.

- **A042 (Zaki)** quantifies it as the [[concept-ai-adoption-gap]] via [[claim-leader-perception-gap]]: 81% of CEOs believe there is a clear AI policy vs. 28% of employees; 76% of execs think the workforce is enthusiastic vs. 31% in reality (all [[entity-bcg-d42|BCG]]).
- **A052 (Hermann/Puntoni/Morewedge)** frames it as the [[claim-adoption-gap]]: 85% of leaders use gen AI vs. only 51% of workers ([[entity-bcg-d52|BCG]] again — the same data house).
- **A078 (Countryman et al.)** localizes it to the factory floor: executive optimism (61% of COOs) vs. frontline fear, driven by [[claim-exec-uncertainty-travels-downstream]].
- **A040 (Deloitte)** documents the behavioral consequence: trust and usage of employer AI actively *declining* even as leaders push harder.

The deepest cross-article insight is A078's [[contrarian-executives-are-also-uncertain]]: the gap is not a communication failure by leaders who have a plan — it exists *because leaders themselves cannot describe the future of work*. That reframes the whole corpus: the fix is not clearer top-down messaging but co-creation (see [[cross-build-with-not-for]]). The gap is also why crude mandates backfire (see [[cross-mandate-tension]]) and why the same tool feels like a copilot to one worker and a cage to another (see [[cross-identity-threat-fobo]]).

**Caution:** the BCG percentages recur across A042 and A052 because they trace to one vendor's survey program — the *direction* is robust, the exact figures are not independent (see [[cross-evidence-quality-caution]]).