---
id: "claim-marginal-business-impact"
type: "claim"
source_timestamps: ["¶5", "¶8"]
tags: ["enterprise-ai", "roi", "business-process"]
related: ["concept-thinkslop", "contrarian-ai-hype-vs-reality", "quote-marginal-benefits", "entity-marc-zao-sanders"]
confidence: "high"
testable: true
speakers: ["Marc Zao-Sanders"]
sources: ["execution"]
sourceVaultSlug: "hbr-seg-execution"
originDay: 8
articleStem: "hbr-cl-77-new-data-using-ai"
sourceUrl: "https://hbr.org/2026/06/new-data-on-how-were-really-using-ai"
sourceTitle: "New Data on How We’re Really Using AI"
---
# AI's current business impact is marginal, not game-changing

**Claim (confidence: high · testable):** AI's current business impact is *marginal, not game-changing*.

According to [[entity-marc-zao-sanders]]'s analysis of **nearly 50,000 records (≈12,637 distinct use cases) from March 2025 to February 2026**, AI is primarily used in the workplace to achieve **'modest, uncontroversial wins.'** More than half of all use cases involve people's jobs — 'work buddy,' enhanced decision-making, career advice — yet there are very few instances where **core business processes are fundamentally rethought or transformed.** Despite the massive hype around generative AI, its actual enterprise deployment currently yields marginal benefits rather than game-changing, systemic transformation. See the verbatim [[quote-marginal-benefits]] and the contrarian framing [[contrarian-ai-hype-vs-reality]].

**Enrichment / nuance:** The claim holds for *typical, broad-based enterprise usage in 2025–2026* but does not capture pockets of more transformative impact or executive expectations. The 2026 *State of AI for Business Report* (2,100+ professionals) finds **74% call AI 'critically' or 'very' important** to near-term success (**39%** 'critically important'), and identifies the main barriers as **human (skills, pace of change)** rather than technical or budget — implying transformational impact may lag due to adoption friction, not inherent limits. Documented **2–4x** gains exist in software engineering, customer support, and marketing. Best framed as *'on average, impact is marginal so far,'* not universally marginal. Connects to [[concept-thinkslop]] — the individual-level flip side of institutions failing to transform.


## Related across articles
- [[claim-95-percent-failure]]
- [[claim-widening-performance-gap]]
- [[claim-translation-difficulty]]
- [[quote-roi-kept-by-employee]]