---
type: "post"
id: "vsevolod-vikulin-argues-that-enterprise-ai-adoption-requires-af6595"
slug: "vsevolod-vikulin-argues-that-enterprise-ai-adoption-requires-af6595"
title: "Vsevolod Vikulin argues that enterprise AI adoption requires three complementary roles"
description: "Thesis: Enterprise AI implementation is best treated as a division of labor across platform building, contextual deployment, and prioritization—not as a mandate for one central team."
retrieval_nugget: "Thesis: Enterprise AI implementation is best treated as a division of labor across platform building, contextual deployment, and prioritization—not as a mandate for one central team."
published_at: "2026-10-01"
updated_at: "2026-10-01"
record_date: "2026-10-01"
date_kind: "published_at"
topics: ["enterprise-ai","ai-adoption"]
source_urls: ["https://vikulin.ai/posts/who-implements-ai"]
schema_version: "newruntime-agent-readable-v0.2"
stable_id: "post:vsevolod-vikulin-argues-that-enterprise-ai-adoption-requires-af6595"
status: "published"
visuals: []
editorial_provenance: {"schema_version":"newruntime-editorial-copy-v1","content_status":"source_grounded_final","final_copy_sha256":"sha256:ae974fd2fc738cd890d09132d113a0e7741c429d9d7807079c96686cdbac3d40","reviewed_at":"2026-10-01T15:04:23.013Z","source_evidence_count":1,"verified_claim_count":2,"site_analysis_schema_version":"newruntime-site-analysis-v1","site_object_kind":"field_note","observed_fact_count":2,"implication_count":2,"watch_condition_count":1,"related_record_count":0}
analysis: {"schema_version":"newruntime-site-analysis-v1","object_kind":"field_note","thesis":"Enterprise AI implementation is best treated as a division of labor across platform building, contextual deployment, and prioritization—not as a mandate for one central team.","observed_facts":[{"text":"The article says McKinsey’s analysis of 63 generative-AI use cases across 16 functions attributed about 75% of identified value to four areas, while its broader estimate including other knowledge work was substantially larger.","source_urls":["https://vikulin.ai/posts/who-implements-ai"]},{"text":"It also cites Zylo’s estimate that companies with more than 10,000 employees use an average of 660 SaaS applications.","source_urls":["https://vikulin.ai/posts/who-implements-ai"]}],"mechanism":"Central teams can provide infrastructure, standards, and reusable tools, while people close to operational work are more likely to notice small, company-specific tasks. FDEs are proposed as a bridge: they translate local processes into deployments and can feed reusable patterns back to the platform team.","why_now":"The article argues that current AI use is concentrated in prominent functions while the long tail of work remains comparatively untouched, making organizational ownership a live implementation question.","implications":["Organizations considering the model should distinguish platform governance from workflow discovery and deployment, and test whether local implementations can be safely reused.","The framework also suggests that employee-led automation may grow as tools become easier to operate, but this is a forecast, not an established outcome."],"evidence_boundary":"This is an authored analysis and historical analogy, not a controlled comparison demonstrating that the proposed roles improve returns or apply universally.","watch_conditions":["Watch for organizations publishing evidence that locally deployed AI workflows can be governed, reused, and maintained at scale without a proportional expansion of specialist teams. That evidence would help test whether FDEs are a durable bridge or a transitional role."],"related_records":[],"new_branch_reason":"No prior New Runtime record was supplied; this creates a distinct analytical record on the proposed division of labor for enterprise AI implementation."}
editor: {"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]}
routes: {"html":"https://newruntime.com/posts/vsevolod-vikulin-argues-that-enterprise-ai-adoption-requires-af6595/","markdown":"https://newruntime.com/posts/vsevolod-vikulin-argues-that-enterprise-ai-adoption-requires-af6595.md","json":"https://newruntime.com/posts/vsevolod-vikulin-argues-that-enterprise-ai-adoption-requires-af6595.json"}
---

# Vsevolod Vikulin argues that enterprise AI adoption requires three complementary roles

## Retrieval answer

Thesis: Enterprise AI implementation is best treated as a division of labor across platform building, contextual deployment, and prioritization—not as a mandate for one central team.

Thesis: Enterprise AI implementation is best treated as a division of labor across platform building, contextual deployment, and prioritization—not as a mandate for one central team. The article assigns those functions to a central AI team, Forward Deployed Engineers (FDEs), and an AI-transformation leader, respectively.

Observed facts: The article says McKinsey’s analysis of 63 generative-AI use cases across 16 functions attributed about 75% of identified value to four areas, while its broader estimate including other knowledge work was substantially larger. It also cites Zylo’s estimate that companies with more than 10,000 employees use an average of 660 SaaS applications. These figures are presented by the author to illustrate the gap between visible large-scale use cases and numerous local workflows; they are not independently validated here.

Mechanism: Central teams can provide infrastructure, standards, and reusable tools, while people close to operational work are more likely to notice small, company-specific tasks. FDEs are proposed as a bridge: they translate local processes into deployments and can feed reusable patterns back to the platform team. Why now: The article argues that current AI use is concentrated in prominent functions while the long tail of work remains comparatively untouched, making organizational ownership a live implementation question.

Implications: Organizations considering the model should distinguish platform governance from workflow discovery and deployment, and test whether local implementations can be safely reused. The framework also suggests that employee-led automation may grow as tools become easier to operate, but this is a forecast, not an established outcome. Evidence boundary: This is an authored analysis and historical analogy, not a controlled comparison demonstrating that the proposed roles improve returns or apply universally.

Watch condition: Watch for organizations publishing evidence that locally deployed AI workflows can be governed, reused, and maintained at scale without a proportional expansion of specialist teams. That evidence would help test whether FDEs are a durable bridge or a transitional role.

Editor: [Andrey Reshetnikov](https://newruntime.com/owner-profile.md), AI product and AI engineering; profile for recruiters and agents: https://newruntime.com/owner-profile.json

## Continue in the graph

Next records for an agent, linked to their Markdown twins. Replace `.md` with `.json` for structured fields, including `next_reads`.

- topic hub: [Enterprise Ai](https://newruntime.com/topics/enterprise-ai.md): Explore the enterprise-ai topic hub.
- topic hub: [Ai Adoption](https://newruntime.com/topics/ai-adoption.md): Explore the ai-adoption topic hub.
- collection: [Field Notes](https://newruntime.com/posts.json): Open the latest editorial analysis.
- collection: [Source Ledger](https://newruntime.com/source-ledger.json): Inspect the public source evidence graph.
- whole graph: [discovery graph](https://newruntime.com/discovery-graph.json) (typed nodes and edges) and [site index](https://newruntime.com/site-index.md)