OpenAI is releasing a new tool that lets you provide “custom instructions” to ChatGPT. In this way, it will be possible to tell Artificial Intelligence what the details are that should always be taken into account when interacting with the user.
In that case, source is in Beta and available for ChatGPT Plus subscribers starting this Thursday (20) — Outside the United Kingdom and the European Union.
Joanne Jang, Product and Behavior Manager of OpenAI, said: Border “if you start a new thread, it will forget everything you talked about in the past. But there are things that can apply to any conversation.”
OpenAI will also ensure that ChatGPT does not violate any company policy. The chatbot may ignore or even not record commands that present some kind of harmful content. It will also be possible to disable the feature via settings.
Possible uses of special instructions
One of the examples given by Jang is its use by teachers. For example, a teacher might include personalized instruction in a third grade classroom. In this case, whenever the tool responds, it will be possible to obtain a language suitable for the students’ age.
Another possibility would be continued use by programmers using a particular language. Answers will always be provided in accordance with predefined instructions.
Anyway, the purpose of the change is to save users from always adding. Long contexts before I ask ChatGPT a question.
While the facility can further complicate interactions with AI, Jang stated that this is one of the reasons the feature rolled out into Beta. First. “We want the model to learn exactly how and when to follow these guidelines.”
Source: Tec Mundo

I am a passionate and hardworking journalist with an eye for detail. I specialize in the field of news reporting, and have been writing for Gadget Onus, a renowned online news site, since 2019. As the author of their Hot News section, I’m proud to be at the forefront of today’s headlines and current affairs.