How do you think adding personality and descriptors to prompts can impact the quality and style of generated text? Have you experimented with these techniques in your own interactions with ChatGPT or similar AI models?
Join the conversation and share your insights with the DEVcommunity!
For further actions, you may consider blocking this person and/or reporting abuse
Top comments (4)
I have experimented with a lot of different techniques and styles for prompting and the final result changes significantly when you ask the AI to act as some professional or write using a style. But the quality also varies if you are using GPT3.5, GPT4, Gemini, CoPilot, or Claude. Also, local LLMs can behave differently, sometimes giving better or worse results.
Overall, making a simple statement (consider you are a "something" with "some time of experience" and writes using "some style") gives better results, especially if you are asking for an analysis/review
Cool.
Yes, I believe that it does so. It tries to reply back in a similar way through tone, words etc.
Definitely, because chat-gpt give the experience of having a company of very professional person who is master in everything even your manager is'nt that professional