Lore

Taxonomy of Corporate AI Points of View by Business Model

Companies' stated points of view on where AI is taking the market diverge sharply, and predictably track each company's own business model — evidence that a market POV (see Point of View on the Market's Future as a Positioning Requirement) is less a prediction than a positioning move:

• Anthropic — AI takes over developer tasks.
• OpenAI — intelligence becomes a metered utility, "like electricity or water."
• Replit — agents will spontaneously generate bespoke tools on demand, replacing SaaS subscriptions.
• Microsoft — models are commodity; the differentiator is owned data surfaces (email, Office, SharePoint) plus "context engineering."
• Salesforce (Benioff) — "our API is the UI": expose everything and let customers build however they want.
• IBM (Krishna) — real-time, open, sovereign AI built on mainframes and orchestration.
• ServiceNow (McDermott) — workflow and governance, not models, are "the whole ballgame."

Each company's stated POV conveniently favors the assets it already owns — exactly what the "why can only we claim this?" credibility test in Point of View on the Market's Future as a Positioning Requirement is designed to catch.

AI-Era Examples by Business Model

Companies' stated futures map directly onto their own core strengths rather than being invented from scratch:

• OpenAI — frames itself as a metered utility: "we're going to be like the electric company and you just buy it from us on a meter."
• Anthropic — frames a B2B, programmer-replacement future, consistent with its enterprise/developer strength.
• Replit — envisions a future with no more SaaS software at all, replaced by generated apps.
• Microsoft — argues models themselves are becoming commodities, and that proprietary context (a company's own data and workflows) is the real moat.
• Salesforce — argues the durable value sits in a "headless" back end — the data, business logic, and workflows underneath whatever front end (chat, agent, UI) customers use.

The pattern: don't reverse-engineer a self-serving future story — identify genuine differentiated strengths first, then extrapolate forward from them. See Point of View on the Market's Future as a Positioning Requirement.

Examples: Utility, Headless Platform, Context Moat, Sun/Planets

Four concrete framings companies use to state their point of view on the AI-disrupted future of their category:

• Electricity/utility metaphor (OpenAI). OpenAI frames its future role as a metered utility — 'the electric company' — rather than an app-specific vendor: 'I don't care, man. I'm just the electric company. As long as I get my bit, that's good.' Signals indifference to what's built on top and a preference for billing by usage over billing by use case.
• Headless Salesforce. Salesforce positions its enduring value as living in the back end — data, logic, and workflows — that many different front ends, including AI agents, can call into: 'look at the value in the back end. All that data, the logic, the workflows, all these things that are very, very difficult to vibe code.' Lets a platform company reposition value away from a commoditized front-end/UI/seat layer toward the harder-to-replicate back end.
• Models-as-commodity / context-as-moat (Microsoft, Satya Nadella). Frames AI models as interchangeable commodities while proprietary business context — documents, SharePoint, email — is the scarce, defensible asset. Companies sitting on proprietary customer/process data can position that context layer, not model quality, as the moat.
• Sun/planets metaphor. A self-centered framing that positions a company's platform as the fixed center other tools and agents orbit: 'we are the sun and all the things are planets that revolve around us.' Used to communicate that ecosystem activity — including new AI agents — still routes through and depends on the platform.

These are all instances of the broader requirement in Point of View on the Market's Future as a Positioning Requirement; the taxonomy is about which asset each company nominates as the durable one once models and front ends commoditize.

Platform-Centrism vs. Indifference: Microsoft and OpenAI

Two philosophies that read as opposed are framed as compatible rather than in conflict: Microsoft's platform-centric position that "everything still hits our stuff" (whatever gets built on top of the AI stack eventually touches Microsoft's platform layer), versus OpenAI's stated indifference to what's built on top — "I don't care, I'm just the electric company." Both can be true simultaneously because they describe different layers of the same stack: one company stakes its point of view on owning the platform surface, the other on being commoditized infrastructure underneath it, and neither claim depends on the other being wrong.

Из тем: Positioning, Marketing, and Go-to-Market