Chollet's argument isn't about AGI or research progress. It's about business and real world impact of LLMs. He outlines two cases: 1. Bull case. LLMs become a

Chollet's argument isn't about AGI or research progress. It's about business and real world impact of LLMs. He outlines two cases: 1. Bull case. LLMs become a kind of general speech/language interface to data. 2. Bear case. LLMs continue to be used primarily to automate customer support agents and help marketing people write blog posts, and within a few years become seen as a blockchain-like bubble: cool tech, not much real world impact. So you're not making an opposing case, because you're responding to a strawman argument that the article isn't about.