The AI boom has been built on a simple assumption: the companies with the best models will capture most of the value. That belief helped push valuations for leading AI developers into the hundreds of billions of dollars and turned model performance benchmarks into headline news.
But markets have a habit of changing the rules. When a technology becomes widely available and customers can switch providers with little friction, price becomes the battleground. New reports suggest that moment may be arriving for artificial intelligence.
OpenAI’s Pricing Move Could Change Everything
According to The Wall Street Journal, OpenAI is considering “drastic price cuts” as competition with Anthropic intensifies. CNBC separately reported that OpenAI is evaluating lower token pricing in anticipation of a broader fight for enterprise customers.
The numbers behind this matter. OpenAI and Anthropic are both spending billions of dollars on infrastructure while racing toward potential IPOs. Yet customers are increasingly scrutinizing AI spending and demanding measurable returns. OpenAI CEO Sam Altman recently acknowledged that AI costs have become “a huge issue” for customers. Uber Technologies (NYSE:UBER | UBER Price Prediction) infamously spent its entire AI budget for the year in just four months.
What’s more interesting is what many enterprise software users are already reporting. Increasingly, companies seem less concerned about whether a vendor uses OpenAI, Anthropic, Google, or another third-party model. They care about results, compliance, workflow integration, and cost.
That’s not what a durable moat looks like. When customers become indifferent to the underlying model, the model starts looking like a commodity.
The Real Moat May Be Customer Data
If model differentiation continues shrinking over the next several years, the winning AI companies won’t necessarily be those with the smartest models. They’ll be the companies that possess the most valuable customer-specific data and can safely deploy AI agents on top of it.
Think about what enterprise customers actually need:
- Regulatory compliance
- Historical transaction records
- Customer relationship data
- Healthcare records
- Supply chain information
- Internal workflow data
The companies already sitting on these datasets are the large enterprise software providers.
Earlier this year, investors punished many software stocks during the so-called “SaaS-pocalypse.” Revenue growth slowed, valuation multiples compressed, and investors shifted attention toward AI infrastructure plays. Yet these software firms maintained control over something far more difficult to replicate than a language model: decades of proprietary customer data.
That data becomes exponentially more valuable when AI agents begin making decisions and executing work on behalf of customers.
Why This Could Create Surprising Winners
Surprisingly, this logic helps explain why some analysts believe companies far removed from the AI model wars could become enormous beneficiaries.
One recent analysis argued that Eli Lilly (NYSE:LLY) could surpass Nvidia (NASDAQ:NVDA) in market value within five years. The reasoning wasn’t that Lilly would build better AI models. It was that proprietary healthcare data, patient relationships, and specialized industry expertise may become more valuable than access to increasingly interchangeable AI engines.
The same principle applies across enterprise software. If OpenAI, Anthropic, Google, and others engage in aggressive pricing competition, the economic value may migrate away from the model providers and toward the companies that own the customer relationship and the underlying data.
In other words, AI models may become the electricity. The software platforms connected to customer data become the power grid.
Key Takeaway
In short, a price war between OpenAI and Anthropic could be a warning sign for investors betting exclusively on model providers. Reports of potential pricing cuts suggest competitive pressure is already building.
Granted, model leadership still matters. The best systems will continue attracting users and enterprise contracts. That said, if switching costs remain low and performance differences narrow, pricing pressure becomes difficult to avoid.
Regardless, investors should pay closer attention to who owns the data rather than who owns the model. The enterprise software companies that many investors abandoned during the SaaS downturn may ultimately control the most valuable asset in the AI economy: trusted access to customer information.
And in investing, the companies everyone hates today often become the stocks everyone wishes they had bought tomorrow.
Top Gaining Stocks
Top Losing Stocks