The AI reset

Turns out even the biggest name in AI has a budget.

OpenAI is scaling back its infrastructure ambitions by more than half, cutting its spending plans from $1.4 trillion down to $600 billion through 2030, according to CNBC. It’s the latest warning sign that future revenue for AI startups may not be able to support the immense cost of expanding their AI systems.

Still, projections remain aggressive: Management expects annual revenue to exceed $280 billion by 2030, a 2,000% increase from last year. To fuel that growth, the ChatGPT maker is closing in on a blockbuster funding round that could see it raise over $100 billion. Nvidia is reportedly weighing an investment of up to $30 billion, alongside strategic backers including SoftBank and Amazon.

The revision reframes expectations for a market that has priced in relentless AI expansion, with hyperscalers collectively guiding toward nearly $700 billion in capital spending this year. Any moderation carries implications beyond AI labs, affecting chipmakers, power suppliers, and data-center developers whose growth assumptions hinge on sustained demand.

It will be particularly interesting to see if Jensen Huang has anything to say about today’s news when Nvidia announces earnings later this week.

The enterprise push

That said, it’s not all doom and gloom. OpenAI is doubling down on enterprise adoption through multi-year alliances with consulting firms Accenture, Boston Consulting Group, Capgemini, and McKinsey. Corporate clients already represent roughly 40% of revenue, and management expects that share to approach 50% by year-end.

These partnerships are designed to help OpenAI gain market share as the company competes with Google and Anthropic. They also support the rollout of Frontier, OpenAI’s new enterprise platform, which functions as a unifying intelligence layer across company systems.

The productivity paradox

Zooming out, Citrini Research argues that the bigger risk isn’t that AI underdelivers—but that it succeeds too well.

In a memo published over the weekend that’s since gone viral, Citrini analysts warn that broad AI adoption could compress margins and weaken competitive advantages across industries. Signs are already visible in software, and as AI becomes embedded across consumer and enterprise activity, the ripple effects could extend far beyond.

The labor market will be central to the shift. With white-collar jobs making up about half of employment, broad automation could cut headcount, with companies pouring the savings back into more AI, creating a positive feedback loop that weakens consumption and traditional growth drivers.

OpenAI and its peers might thrive in that world. For the rest of us, it might be time to update our resumes.—SY