OpenAI Altered Reported Metrics for GPT-6-Astra

Jeremy Kahn, Fortune's AI Editor, posted that OpenAI quietly altered performance metrics for GPT-6-Astra in ways that favored the model. He added that the company continues to revise other metrics after launch. Kahn credited reporting by Emily Forlini and included a link to her Fortune article, which covers the adjustments and notes a rare delay in the publication of the evaluation results.