Claude: enterprise and model leverage lens
Compute matters, but the differentiator shifts to what companies do with it: enterprise revenue, coding agents, and model leverage.
A ranking of which AI companies are best positioned to own the next phase: compute, enterprise deployment, platform control, consumer habit, model leverage, momentum, brand, wedge, and X-factor.
The framework scores who can turn intelligence into infrastructure, products, habits, and business outcomes at global scale. Compute carries the most weight; enterprise and platform control determine whether AI becomes durable value.
Physical scale, energy, chips, clouds, and cost structure.
Procurement trust, governance, integrations, and seat expansion.
The layer where developers, agents, and businesses build.
Default surfaces, habits, feedback loops, and mass adoption.
Model strength turned into useful products, agents, and economics.
Release cadence, customer wins, talent, growth, and execution speed.
Trust, category leadership, inevitability, and market story.
The unique asset or entry point others cannot easily copy.
Founder aggression, regulatory position, data, culture, or surprise.
Each model keeps the same basic categories, then shifts the weights toward a different view of where AI company advantage comes from.
Compute matters, but the differentiator shifts to what companies do with it: enterprise revenue, coding agents, and model leverage.
Reduce compute and vibes categories, then reward model quality, product cadence, and agent execution.
Slightly boost enterprise, platform, and momentum because agents need governance, identity, and a runtime.
Hardware supremacy and B2B integration dominate; consumer chat is treated as a high-volume loss leader.
Claude wants to pull the lens toward enterprise revenue and model leverage. ChatGPT wants model quality and momentum to matter more. Grok wants agent runtime control. Gemini makes the hardest industrial argument: compute ownership and B2B integration beat consumer pull.
The consensus calculation uses complete AI scorecards only: Claude, ChatGPT, Grok, and the updated Gemini/Google report.
| Company | AI avg | AI range | Rank spread | NLW total | NLW vs AI avg |
|---|---|---|---|---|---|
| 91.4 | 89-93.5 | 0 | 74 | -17.4 | |
| OpenAI | 85.4 | 80-92.5 | 1 | 74 | -11.4 |
| Microsoft | 84.9 | 81-89 | 2 | 62 | -22.9 |
| Anthropic | 83.1 | 78-86 | 2 | 70 | -13.1 |
| Amazon | 80.4 | 76-85.5 | 2 | 64 | -16.4 |
| Meta | 75.9 | 67-82 | 1 | 60 | -15.9 |
| xAI | 66.9 | 54-75 | 0 | 61 | -5.9 |
| Apple | 61.3 | 51-67 | 0 | 58 | -3.3 |
Each answer reveals a different theory of power: full-stack control, enterprise lock-in, consumer gravity, model leverage, or industrial scale.
Google is most complete, but Anthropic is the fastest-improving challenger on the enterprise dimensions.
Most complete AI company on the board.
Perfect model-leverage and momentum scores in Claude's framing.
Still the consumer king, but infrastructure and governance cracks matter.
Highest enterprise score, with less AI-native consumer pull.
The Switzerland of AI infrastructure.
Glasses and social reach are real, but model narrative is shaky.
Maximum X-factor, minimum enterprise trust.
Massive device wedge, weak hyperscaler posture.
Google barely edges OpenAI because the rubric rewards full-stack power across compute, cloud, enterprise, platform, and consumer surfaces.
Best full-stack AI position.
Strongest AI-native brand and consumer interface.
Still the enterprise AI default.
Hottest model-and-enterprise momentum story.
Infrastructure giant with less first-party AI magic.
Massive reach, open-model influence, glasses, and weak enterprise.
High-velocity outsider, still unproven in enterprise trust.
Best device wedge, weakest current AI execution story.
The companies that own the layer where agents run inside companies are pulling ahead.
End-to-end ownership of frontier models, TPUs, and default consumer surfaces.
Deepest enterprise moat in tech.
Default AI interface plus serious infrastructure ambition.
Multi-cloud frontier lab with enterprise pull.
Compute, social distribution, open models, and glasses.
Neutral hyperscale leader with Bedrock and Trainium.
Execution speed, Colossus, X data, and maximum X-factor.
Device wedge plus privacy architecture, but not a hyperscale compute race participant.
The race has moved from algorithmic iteration to industrial mobilization, where physical hardware and enterprise integration decide value capture.
The most well-rounded competitor through vertical hardware integration and Gemini ecosystem reach.
Apex enterprise software predator with massive infrastructure ambition.
Highest consumer mindshare but structurally dependent on external infrastructure.
Preferred enterprise foundation model provider with strong trust economics.
Open-source ecosystem control and social app distribution.
Custom silicon and enterprise neutrality keep AWS in the fight.
Extreme execution speed and Colossus compute ambition.
Device ubiquity offset by algorithmic delays.
The ranking used throughout the comparison.
| Company | Compute20 | Enterprise15 | Platform15 | Consumer10 | Model10 | Momentum10 | Brand10 | Wedge5 | X-factor5 | Total |
|---|---|---|---|---|---|---|---|---|---|---|
| 17 | 8 | 13 | 9 | 7 | 3 | 9 | 5 | 3 | 74 | |
| OpenAI | 12 | 10 | 11 | 9 | 9 | 10 | 7 | 4 | 2 | 74 |
| Anthropic | 10 | 14 | 10 | 4 | 9 | 8 | 9 | 3 | 3 | 70 |
| Amazon | 16 | 12 | 10 | 2 | 5 | 6 | 6 | 4 | 3 | 64 |
| Microsoft | 14 | 14 | 12 | 2 | 5 | 3 | 6 | 5 | 1 | 62 |
| xAI | 16 | 6 | 9 | 5 | 5 | 2 | 6 | 4 | 8 | 61 |
| Meta | 13 | 5 | 12 | 9 | 3 | 4 | 4 | 5 | 5 | 60 |
| Apple | 5 | 7 | 13 | 7 | 5 | 3 | 8 | 5 | 5 | 58 |
NLW is the toughest judge in the room: lower totals, sharper gaps, and less incumbent benefit.
Google and OpenAI tie at 74, but for very different reasons: Google is breadth; OpenAI is momentum and AI-native habit.
Anthropic is the clean enterprise challenger and the company most rewarded by alternate methodologies that raise enterprise and model leverage.
Microsoft and Amazon look stronger in AI-generated rankings because the models reward infrastructure and enterprise incumbency more heavily.
Apple lands last because the device wedge is real, but current AI execution still trails the field.
Join ... community members ranking the field alongside NLW and the AI consensus.
| Company | Compute20 | Enterprise15 | Platform15 | Consumer10 | Model10 | Momentum10 | Brand10 | Wedge5 | X-factor5 | Total |
|---|---|---|---|---|---|---|---|---|---|---|
| 0 | ||||||||||
| OpenAI | 0 | |||||||||
| Anthropic | 0 | |||||||||
| Amazon | 0 | |||||||||
| Microsoft | 0 | |||||||||
| xAI | 0 | |||||||||
| Meta | 0 | |||||||||
| Apple | 0 |