OpenAI recently announced they would not be releasing their GPT-2 language model due to concerns about potential negative effects. Some argue that this decision was unnecessary as the trajectory of language modeling is clearly heading towards scaling up models and improving performance.