4 Simple Steps to Transform Your AI Prompting from Average to Exceptional
The more I use AI the more I have noticed a consistent patterns: monotonous tone, repetitive phrasing, generic or over-polished language, lack of emotion, etc. And often the output feels ‘average’.
And that is no surprise. AI has been trained on a vast amount of human-generated content, a mix of high and low quality data samples, a blend of brilliance and banality. As a result, the output tends to reflect the middle ground. In other words, AI is optimized for average, conventional wisdom.
Sometimes, that is fine. An average request deserves an average response.
But here comes the problem. We want AI to, not only save us time, but also to produce content that is better than average, unique and high quality. And average will not cut it.
So how can we improve our prompting to get ‘better-than-average results’?
Here are my 4 simple but essential steps to improve your AI prompting technique:
-
Assign AI Role
Giving the AI a specific persona sets expectations and narrows the response style. As AI to act as a ‘Technical Copywriter, or ‘UX Designer’ will help to shape tone of voice and depth, providing tailored and more relevant responses. -
Define a clear Task
Be precise in what you are asking for. The clearer the task, the sharper the response will be. Instead of saying ‘Help me write blog post’, be specific: ‘Draft a blog post on how to improve prompting technique for higher quality output’. - See what I did here👏 -
Provide additional Context
Context and scope is everything. Describe what you are trying to achieve, who is the audience, why it is important. Including docs (brand guidelines, research, manuals, etc.) will help LLM to better ‘understand’ you needs and improve response accuracy. -
Specify Output Requirements
Length, format, tone or structure - you need to be specific on what you want the output to be. Do you want a concise summary, a detailed analysis, or a list of key takeaways?
Considering AI is here to stay, the ability to prompt effectively, although hard to master, is a skill too important to ignore.
I hope you find above as useful as I have 😎