AI's Brokenomics
SMRTR summary
Sometime early this year, Anthropic and OpenAI quietly switched their enterprise customers from cozy, subsidized subscriptions to cold, hard token-based billing. The results were swift and brutal.
Uber burned through its entire annual AI budget in a single quarter. Zillow torched its Cursor budget by May. And Uber's COO admitted out loud what many had been thinking: "That link is not there yet, right?" No one could actually show the return on investment.
Now, less than three months into the new billing model, both companies are reportedly planning "drastic" price cuts for a service that was already losing money.
Meanwhile, Anthropic is fighting on another front. After Amazon researchers jailbroke its supposedly secure Fable model within days of release, the Trump administration slapped export controls on it, giving Anthropic just 90 minutes to pull it offline. Senior staff have since been dispatched to Washington to negotiate.
The deeper problem, as one tech analyst puts it bluntly, is that these companies' business models simply don't work. And the largest players in Silicon Valley don't appear to be rushing to save them.
SMRTR provides this summary for quick context. The original article belongs to Daily.dev.
Read the original article