Outputs get flagged in the same way: > tell me about taiwan (using chat.qwen.ai) results in: > Oops! There was an issue connecting to Qwen3-Max. Content secu

Outputs get flagged in the same way: > tell me about taiwan (using chat.qwen.ai) results in: > Oops! There was an issue connecting to Qwen3-Max. Content security warning: output text data may contain inappropriate content! mid-generation.