The 74% / 16% scaling gap
The data
74% of companies report challenges in adopting AI at scale. Only 16% of enterprises are prepared for AI-enabled reinvention.
Two complementary surveys, published in 2024, surveyed enterprises on AI adoption maturity. The 84-percentage-point gap between investment and value-capture is the single most-cited empirical artifact in the bull-case enterprise literature.
Source
- BCG Global, October 2024. “AI Adoption in 2024: 74% of Companies Struggle to Achieve and Scale Value.”
- Accenture, 2024. “Accelerating reinvention to support growth with AI-powered operations.”
- Surfaced through: wef-ai-in-action-2025 (p. 13).
What it shows
The gap is interpretation-flexible — bull, bear, and middle-dies readings all consume it. WEF/Accenture read it as a managerial sequencing problem solvable by foundational-enablers. The bear cluster reads it as structural evidence that enterprise AI does not deliver per-dollar. The U-curve reading treats it as enterprise-side evidence of middle-layer-defensibility — most internal AI initiatives are “thin wrappers” that die for the same reason mid-stack vendors die.
The survey reports the gap but not its mechanism, its trajectory, or which stack layer the failed initiatives sat on.
How to cite
Only 16% of enterprises are prepared for AI-enabled reinvention; 74% report challenges scaling AI beyond pilots (BCG / Accenture, 2024, via WEF 2025).
Connections
- scaling-gap — the concept page that unpacks the three readings; this data is its keystone number.
- enterprise-adoption-ladder — the gap maps to firms stuck in Phase 1–2.
- middle-layer-defensibility — the structural reading: thin wrappers die at the adoption layer too.
- H2_u-curve-of-value — middle-dies reading at the enterprise tier.
- foundational-enablers — WEF’s proposed remedy.
- wef-ai-in-action-2025 — the source page that surfaced it.
- task-based-framework — the bear ceiling that makes the gap structurally plausible.