Others might pursue a smaller model that works as well as a larger model. If that can be done, whoever does it can very effectively compete on price/performance

Others might pursue a smaller model that works as well as a larger model. If that can be done, whoever does it can very effectively compete on price/performance. It seems that to assume otherwise (the only way to improve is to get bigger) is to assume that OpenAI already has found the optimal architecture. That seems unlikely.