The sudden unavailability of a top-tier proprietary AI model reveals a critical business risk. Enterprises now see open-source models, run on local hardware, not just as a cost-saver but as a necessary strategy for predictable access and business continuity.
Instead of relying on one powerful model for all tasks, the leading strategy is 'smart routing'—using a panel of models and directing each task to the most appropriate one. This compound architecture demonstrably beats single frontier models on both cost and performance.
Contrary to expectations of falling AI costs, the move from simple chatbots to complex, multi-step agentic systems is causing an explosion in token usage. A single user can trigger hundreds of agents, making expensive frontier models economically unsustainable for many application-layer companies.
A major contradiction in US policy has emerged: while the government bans allies from top US AI models over security concerns, Microsoft is preparing to integrate a Chinese-developed open-source model into the core productivity stack used by America's largest corporations.
In the vacuum left by banned US frontier models, Chinese labs are releasing powerful and cost-effective open-source alternatives like ZAI's GLM 5.2. These models are proving competitive on valuable, complex tasks like UI design and coding, but at a fraction of the cost.
The US government's ban on a frontier AI model ("Fable") caused European allies at the G7 summit to pivot from discussing a united front against China to pleading for access and expressing concern over the US government's control over critical AI technology.