Hi OpenAI Team and Community, Over the past two days, I have encountered a sudden and severe spike in safety refusals that is actively blocking my academic research workflow. It appears a recent update to the safety classifier has become overly aggressive…