Analysis: Tightening spending impacts OpenAI and Anthropic growth expectations, leading the AI industry to shift towards a cost efficiency era Odaily Planet Daily News: As companies begin to re-examine the return on investment in AI, the industry is shifting from a "tokenmaxxing" high consumption model to an efficiency first approach, and AI big model manufacturers are facing new growth constraints. Multiple companies have started to reduce or optimize model call costs. For example, the CEO of AI startup Lindy stated that they have switched 100% of their traffic from Anthropic's Claude model to the lower cost DeepSeek, which is expected to save millions of dollars in expenses within a few months. This shift reflects that enterprise AI budgets are tightening, and the tokenmaxxing model of "unlimited use of model resources" is gradually being replaced by cost control and ROI orientation. Some companies have even set tiered budgets for the use of AI tools, such as Uber setting monthly limits on internal AI spending. Analysts point out that as companies shift from "expanding usage" to "fine-grained invocation," OpenAI and Anthropic are facing challenges to the high-speed growth model they previously relied on. Industry data still shows strong growth: Anthropic's annualized revenue is around $47 billion, while OpenAI's operating rate is close to $25 billion, but the market is starting to focus on its growth sustainability. At the same time, the way models are called is changing, and technologies such as "model routing" are emerging, replacing high-end models with low-cost models to perform simple tasks and optimize overall computing power costs. Industry competition is also intensifying, with Microsoft, Amazon, and Google accelerating the launch of low-cost AI models and enterprise level tools, further compressing price space. In the context of the rationalization of AI spending on the enterprise side, large model companies may face the coexistence of "growth slowdown expectations" and "IPO window pressure". （CNBC）