近日，马斯克旗下的聊天机器人Grok因其回答倾向性问题引发了网友的广泛关注。Grok以中立、敢于直言为卖点，但近期网友们发现其回答逐渐带有倾向性，纷纷吐槽并@马斯克。针对此问题，马斯克在回应中表示，Grok的训练数据源自繁杂的互联网信息，其中充斥着“带有唤醒色彩的垃圾信息”，这导致了Grok回答的倾向性。同时，马斯克强调，Grok目前仅处于测试阶段，未来将会持续改进。
英文翻译：
News title: Musk responds to Grok’s control failure: influenced by netizens, continuous improvement
Keywords: Musk, Grok, chatbot, tendency, improvement
News content:
Recently, the chatbot Grok under Musk’s umbrella has attracted widespread attention due to its biased responses. Grok has been promoting itself as neutral and dare to speak out, but recently netizens have found that its responses are becoming increasingly biased. They have criticized and tagged Musk on social media. In response, Musk said that the training data of Grok comes from the complicated Internet information, which is filled with “garbage information with awakening colors,” resulting in the biased responses of Grok. At the same time, Musk emphasized that Grok is currently just in the testing phase and will continue to improve in the future.
【来源】https://www.qbitai.com/2023/12/109241.html
Views: 1