It's funny you say that, I thought this would be an article about how Anthropic have managed to produce a better (coding) product than OpenAI despite having 1/10th of the funding.
The new versions of Opus (4.5 and 4.6) are absolutely amazing - first time I've felt it necessary to throw hundreds of dollars in a single month at Cursor.
I heard similar things about the older models too (Sonnet 3.5 beating GPT-4 etc.) but sadly only jumped on the Cursor train in the last 12 months or so.
The problem is not the models, is the moat and budget. Google and X still have money and are profitable, all the other AI companies are losing billions per year.
And customers will happily switch from one model to another in a heartbeat.