---
id: "claim-genai-hardest-to-value"
type: "claim"
source_timestamps: ["§ A Survey of Executives Suggests Anticipatory Effects"]
tags: ["roi", "generative-ai"]
related: ["concept-ai-economic-value-measurement"]
confidence: "high"
testable: true
speakers: ["Thomas H. Davenport", "Laks Srinivasan"]
sources: ["execution"]
sourceVaultSlug: "hbr-seg-execution"
originDay: 8
articleStem: "hbr-foci-62-layoffs-ai-potential-not-performance"
sourceUrl: "https://hbr.org/2026/01/companies-are-laying-off-workers-because-of-ais-potential-not-its-performance"
sourceTitle: "Companies Are Laying Off Workers Because of AI’s Potential—Not Its Performance"
---
# Generative AI Is the Most Difficult Form of AI to Assess for Economic Value

**Claim (confidence: high · testable: true):** According to the December 2025 survey of 1,006 executives, **44% identified generative AI as the most difficult form of AI technology to assess for economic value** — harder than analytical AI, deterministic AI, and agentic AI.

This difficulty contributes directly to the disconnect between the high expectations of executives and the lack of *actual*, performance-based headcount reductions: you cannot easily justify displacement by a value you cannot measure. Grounded in [[concept-ai-economic-value-measurement]]; the AI typology it assumes is covered by [[prereq-ai-typology]]; its ironic framing is [[contrarian-genai-hardest-to-value]].

**Enrichment corroboration:** Grant Thornton's *AI proof gap* (78% not confident they could pass an AI governance audit in 90 days) and EY's transformational-results gap (only 28%) both support the view that generative AI value is uniquely hard to demonstrate and defend.


## Related across articles
- [[question-defining-ai-roi]]
- [[claim-marginal-business-impact]]
- [[concept-ai-economic-value-measurement]]