AI in the Enterprise

Strategic Center Evaluation

Adjacent Publisher Context

The following materials are published by OpenAI and address topics relevant to the Five Forces, but are not explicitly positioned as components of the AI in the Enterprise guide. They are not included in the /25 coverage score above.

Narrative Strengths & Gaps

Internal Pressures & Capabilities. This is the guide’s centre of gravity. It moves from OpenAI’s own Research–Applied–Deployment structure and iterative deployment model to six internal disciplines: systematic evals (Morgan Stanley’s translation, summarisation and expert-comparison evals), early investment and broad adoption (90% of Klarna employees using AI daily), fine-tuning on proprietary data (Lowe’s), expert-led adoption (BBVA’s 2,900+ custom GPTs created with Legal, Compliance and IT Security involvement), developer platforms (Mercado Libre’s Verdi) and bold automation goals. The treatment spans strategy, operations and risk, with measurable outcomes attached to each lesson. Guidance stays at the level of principles and worked examples rather than stage-gates or templates. The guide does not assess where an organisation stands today (Map: Not met). Its seven lessons and iterative approach point the way forward without sequencing them into a route (Course: Partly met). Its destination is implicit: an enterprise where AI raises workforce performance, automates routine operations and powers products, with the reader choosing which use cases and automation goals to pursue.

Global & Geopolitical Dynamics. The guide does not address this force. Its case studies span several regions (Mercado Libre’s adaptation to Spanish and Portuguese dialects, for example), but cross-border data, sovereignty and national AI policy fall outside its scope. That is consistent with its focus on lessons that “apply across all markets”.

Societal & Regulatory Pressures. The guide builds accountability into its first lesson: evals test whether outputs are “more compliant? Safer?”, and the conclusion rests success on “rigorous evaluations, and safety guardrails”. BBVA’s adoption programme works with Legal and Compliance “to ensure responsible use”, and the closing security-and-privacy page sets out vendor commitments (no training on customer content, SOC 2 Type 2 and CSA STAR Level 1 alignment, access controls, flexible retention). The eval discipline gives leaders a directional practice for checking compliance and safety as they scale. Governance structures, regulatory obligations and public expectations are not treated as a strategic dimension of their own.

Industry & Competitive Pressures. The market-facing dimension appears in Lesson 2, which positions AI inside core products to “create new customer experiences”, and in Lesson 3’s case that value compounds for those who start early. Indeed’s leadership frames this as infrastructure that “will help us grow revenue”, and Klarna’s as “better returns for our investors”. The guide offers directional advice (embed AI in products, start early) without analysing competitors, ecosystem dynamics or vendor positioning, which remain outside its scope.

Human Needs & Aspirations. The guide gives this external force a dedicated treatment through Indeed’s story, framed as “humanizing job matching”. It argues AI can make customer experiences “feel more human because they’re more relevant and personalized”, and that explaining “why” a job fits is “a profoundly human outcome”. Klarna’s results are reported with satisfaction scores “on par with human support”. The guidance is directional: explain, personalise and measure satisfaction. Broader public sentiment, such as attitudes to displacement and citizens’ aspirations for AI, falls outside the guide’s customer-centred lens.

What this framework does not give you. A map of where the organisation stands today. Without a clear picture of where it stands, an enterprise cannot tell which moves matter most. A sequenced route from today to the destination; the seven lessons are presented as parallel disciplines. Without a route, even a well-chosen destination stays an aspiration. A way to choose your own destination beyond selecting use cases and automation goals; it supports an enterprise where AI raises workforce performance, automates routine operations and powers products. An enterprise that has not stated its own ambition cannot check whether this destination is the right one for it. Global & Geopolitical Dynamics. Sovereignty, cross-border data rules and export controls can change which AI an enterprise can use, and where.

Synthesis. AI in the Enterprise serves executives and transformation leads at organisations starting or scaling AI adoption who want proven patterns from large deployments, especially leaders who need evidence that evals, expert-led adoption and developer platforms deliver measurable returns. Its destination fits organisations whose ambition is broad, use-case-led adoption across workforce, operations and products. Its explicit separation of augmentation and automation is a useful anchor for the Strategic Center. The Five Forces lens adds a picture of where the organisation stands, a sequenced route, a way to set the destination itself, and the external dimensions the guide leaves aside: geopolitical and sovereignty exposure, regulation as a strategic dimension, competitive and ecosystem positioning, and a declared time horizon.

Sources

Tier 1 (scored), recency-gate pass unless noted:

1. AI in the Enterprise: Lessons from seven frontier companies – April 2025 (file last modified 11 April 2025); 25-page guide; within the 24-month window until April 2027

Tier 2 (reported in Section 2.75, not scored):

1. Identifying and scaling AI use cases – April 2025 (file last modified 11 April 2025)
2. A practical guide to building agents – April 2025 (file last modified 11 April 2025)
3. Staying ahead in the age of AI – December 2025 per Step 1 research; no date shown on the page; not linked from the guide, classified Tier 2

Tier 3 (third-party corroboration, not scored):

1. No Tier 3 sources are cited in the body of this assessment.

Excluded on recency or quality (not scored):

1. No sources were excluded on recency or quality. A May 2026 OpenAI page on how enterprises are scaling AI, found in Step 1 research, could not be opened and is not cited.