July 25, 2026
AI budget bubble goes pop
Corporate America Has Suddenly Decided to Stop Blowing Money on AI
Bosses finally asked why they were paying champagne prices for robot helpers
TLDR: Companies are cutting AI costs by using a mix of cheaper tools instead of relying only on the biggest, priciest providers. In the comments, people cheered the common-sense savings, mocked past AI overspending, and argued that cheaper Chinese options may be the real shake-up.
Corporate America’s latest plot twist? After spending like there was no tomorrow on artificial intelligence, companies are suddenly discovering they can shop cheaper. The Wall Street Journal says businesses are mixing and matching different AI systems instead of blindly paying top dollar to the biggest names. Translation for normal people: instead of hiring the fanciest robot brain for every task, they’re using the expensive one only when it’s really needed and grabbing a budget option for the boring stuff.
And the comments were absolutely ready for this reality check. One startup worker basically shrugged and said, yes, of course they mix models, but not just to save money — it’s about getting different strengths from different tools. Another user flexed that they spend less than 40 euros a month and still get 10 times more work done, which is the kind of comment guaranteed to make enterprise sales teams break into a cold sweat.
Then came the real tabloid gold: people mocking the old “spend more tokens, get more magic” mindset. One commenter roasted elite AI hype with a surreal joke about putting LSD in the water at Davos so Sam Altman can keep his infinity pool. Another called out the once-fashionable craze of “tokenmaxxing” like it was a cursed internet trend everyone now pretends they never supported. And yes, there was drama over Chinese models too, with one commenter saying nearby hosted Chinese systems are cheaper, secure, and less likely to suddenly refuse during a coding session. Translation: the community mood is shifting from AI FOMO to show me the receipt.
Key Points
- •The article says U.S. companies are reducing AI spending by adopting lower-cost models.
- •Some businesses are using AI models built in China as part of that cost-cutting shift.
- •Companies are increasingly combining multiple models rather than relying only on OpenAI or Anthropic.
- •This multi-model approach is described as shopping "a la carte" for AI based on task and price.
- •Cursor is highlighted as a potential beneficiary because its software works across different AI models.