It's not quite enterprise ready and yet it allegedly has >40% market share. https://mlq.ai/news/anthropic-captures-majority-share-of-new... Just imagine how many organizations will be willing to adopt it once it gets to 3 nines?

And yet they and their proponents would have us believe that their LLMs are capable of producing high quality code with little supervision. Perhaps they should try eating their own dogfood first?