Community debates overblown distillation accusations against open models

A Reddit post argues that accusations of open models being distilled from closed APIs like GPT-4 or Claude are overblown, citing differences between training on outputs and true distillation. The post highlights a recurring cycle where strong open models face immediate skepticism.