Microsoft Supported OpenAI to Enable ChatGPT User Customization
OpenAI, the San Francisco-based business that created the ChatGPT chatbot, announced on Thursday that it is working on an updated version of the popular chatbot that users may customize. This change has been made to allay concerns regarding political as well as other biases while also allowing for a wider range of viewpoints.
According to a blog post, “this will require permitting system results that other people (including ourselves) may strongly disagree with,” and it suggested customization as a solution. Nonetheless, “some constraints on system behavior will always exist.”
Several US media sites warned earlier this week that the results from Microsoft’s latest Bing web browser, which is also driven by OpenAI, may be harmful.
According to OpenAI, ChatGPT’s responses are initially trained on big-text datasets that are accessible online. In the second phase, humans analyze a smaller dataset and provide recommendations on what to do in various circumstances.
The human inspector should instruct ChatGPT to respond “I can’t answer that” if a user requests adult or hate speech-related content.
