Why KVARK
Why enterprise AI stalls
Where AI projects get stuck on the way to measurable value, and how KVARK helps.
Where AI projects get stuck, and how KVARK helps.
What the research says
Gartner expects over 40% of agentic AI projects to be cancelled before the end of 2027, because costs rise, business value stays unclear or risk controls fall short. The figure is a prediction, made in June 2025, rather than a measured rate.
Gartner expects over 40% of agentic AI projects to be cancelled before the end of 2027.
Gartner, 25 June 2025 Source 1
In McKinsey’s survey, 80% of respondents report better individual productivity from AI, while 37% credit AI with some EBIT impact at enterprise level. Gains for individuals do not, on their own, add up to financial results for the whole organisation.
In McKinsey’s survey, 80% of respondents report better individual productivity, while 37% credit AI with some enterprise-level EBIT impact.
McKinsey, August 2026 Source 2
In the same survey, about 20% of respondents say the cost of running AI, tokens included, limits how much their organisation uses it. Like the two figures above, this is what respondents reported; the outcomes were not audited.
In the same survey, about 20% of respondents say running costs, tokens included, limit AI use.
McKinsey, August 2026 Source 2
Six barriers and how KVARK helps
About this assessment
This mapping is Egzakta’s assessment, and no analyst has endorsed it. KVARK tackles technical and operating barriers. Value also depends on which use cases you pick, how you integrate them and redesign work, and whether employees adopt them: 83% of CEOs in IBM’s 2026 survey say AI success depends more on adoption than technology.This mapping is Egzakta’s assessment, and no analyst has endorsed it. Value also depends on use cases, integration, work redesign and adoption: 83% of CEOs in IBM’s 2026 survey put adoption above technology for AI success. Source 3
Sources
- Back to reference 1
Gartner: agentic AI project cancellation forecast, 25 June 2025.
- Back to reference 2
McKinsey: The state of AI in 2026, 25 August 2026. Self-reported by respondents and not audited.
- Back to reference 3
IBM: CEOs are reshaping C-suite roles for the AI era, 4 May 2026. IBM asked 2,000 CEOs and leaders in equivalent roles across 33 geographies.
Start with one use case. Run it under your rules.
We show KVARK on your kind of documents, your access rules and your deployment constraints. Most teams start with a four to six week analysis and a six week pilot.
We show KVARK on your kind of documents, access rules and deployment constraints.