2023 年底，《纽约时报》指控 OpenAI 盗用其数百万篇文章用于 AI 训练。现在 OpenAI 公司回应称，《纽约时报》的诉讼毫无根据。这家人工智能公司表示《纽约时报》” 没有完整地讲述故事”。OpenAI 认为，引起争议的内容源自于” 多年前的文章”，这些文章散布在各个网站上。该公司还说，似乎有人” 故意操纵了提示”。在重新生成提示时，模型的反应并不像《纽约时报》所暗示的那样，《纽约时报》有选择性地选择了支持该报主张的例子，也被称为 Cherry picking（挑樱桃，指有选择地挑选出符合自己观点或论点的事例或证据，而忽视或排除不支持自己观点或论点的事例或证据）。
At the end of 2023, The New York Times accused OpenAI of using millions of its articles for AI training without permission. OpenAI has now responded, stating that the lawsuit by The New York Times is baseless. The artificial intelligence company claims that The New York Times did not “tell the whole story.” OpenAI believes that the controversial content in question originated from “articles from years ago” that were scattered across various websites. The company also alleges that someone may have “deliberately manipulated the prompts.” When regenerating prompts, the model’s response was not as implied by The New York Times. The newspaper selectively chose examples that supported its own claims, a practice commonly known as cherry picking.
微信扫一扫