If the hyperscalers needed to wipe out most of their income on interest expense they’d lose a large amount of their market capitalization. This could drop the stock market a huge amount, and a lot of spending is driven by the “wealth effect” of households feeling wealthy.

Headline doesn't really match the facts in the article. The article seems to say "hyperscalers are borrowing an enormous amount and so far people are lending to them. Other people are worried that this will stop".

Police who seed crime? (Surely.)

Developers who insert bugs?

Saying routers don't work is sort of like saying serverless doesn't work. It depends on when and how!

One routing implementation that recently launched here is interesting (https://news.ycombinator.com/item?id=49099143). It routes based on the models' initial trajectories. This is like having multiple developers get started, seeing what they're doing, then pulling all but one off the project. It should work, but doesn't seem ideal!

I've gotten routing working well for typical chatbot prompts in http://pellmell.ai. This is fine because prompts are easy to classify into category buckets (for example: legal, medical, general knowledge, code). And models definitely have strengths and weaknesses. You want Gemini to answer General Knowledge and you want Claude to answer coding.

I did read the blog post and I’m not sure what you mean by your reply. I don’t think the blog post addresses my comment.

I'm skeptical though. In order to pick which model is on the right trajectory, you actually need intelligence. But real intelligence would make your system painfully slow and more expensive. I suspect you're using a classifier of some sort, but I also suspect what it's really measuring is confidence.

Most likely, this is a fantastic approach for the kind of problem where there's uncertainty but only one correct solution. But this is going to be really bad for cases where there are many potential solutions, some of which look good but are in fact bad. You only show one benchmark, and I'm wondering if it happens to be nicely shaped for this kind of router. Have you run it on DeepSWE?

The strange thing is that I somehow convinced myself that I wasn’t colorblind during my childhood, in spite of obvious evidence, because I often could figure out colors. Then during an eye exam at 19 years old I failed 11/12 vision tests.