Director, Learning Design & Development| PMIAsheville, NC, United States
With Generative AI, iteratively refining and optimizing prompts can lead to better AI-generated results. This may involve adjusting the specificity or clarity of the prompt to increase relevance and accuracy of results.
What examples do you have of how improving a prompt drastically changed the output quality? What specific changes did you make that led to the improvement?
Refining a prompt can completely change the quality of the output. Adding context, defining the desired format, specifying the audience, and including constraints helps the AI generate responses that are more relevant, accurate, and useful. I've found that prompt iteration is one of the most effective ways to improve GenAI results.
Saving Changes...
Anonymous
Ai is a very nice and helpful tool Saving Changes...
Oh boy, I have learned a lot about GenAI. I will ensure I keep asking questions or refining to get the correct output I will keep BAB, RTF, TAG in mind
Theola DuBoseFunctional Manager| BT AmericasLilburn, Ga, United States
Using Microsoft Copilot has significantly improved the quality of my project management deliverables by increasing both efficiency and consistency. It enables me to quickly draft professional communications, status reports, meeting summaries, risk updates, and stakeholder presentations while maintaining a clear and polished tone. By helping organize complex information, identify key themes, and tailor content for different audiences, Copilot allows me to produce more concise, accurate, and actionable outputs. It also supports better decision-making by helping analyze information, highlight risks, and surface important details that might otherwise be overlooked. Most importantly, Copilot reduces the time spent on administrative tasks, allowing me to focus more on stakeholder engagement, risk management, strategic planning, and overall project execution, ultimately improving both the quality and value of my work. Saving Changes...
I've found that prompt refinement makes one of the biggest differences in GenAI output quality. The more specific I am about the objective, context, audience, format, and constraints, the more accurate and useful the response becomes. Rather than accepting the first answer, I treat prompting as an iterative process, refining my instructions until the output aligns with my intended goal. This approach consistently produces more relevant and actionable results. Saving Changes...
Anonymous
Refining a prompt eliminates a lot of noise, the response is more focused and aligned with the desired output. In my experience, providing an example/image is tremendously helpful to clarify and confusion that the AI may have trouble interpreting in the initial prompt request.
Simply stated - better questions get better answers. The techniques and approaches presented in this class remind me that minutes or hours spent thinking through prompts can save more hours and days in later iterations when the LLM doesn't provide what we want. Saving Changes...
I use AI to get SQL language for data analysis and in one example, the results were improvedby breaking down the task into smaller sub tasks. I wanted a query that would help determine reoccurrence across several years of data and without specifying the additional context that the dataset had several categories, the result only focused on looking at reoccurrence over time, ignoring the cross sectional analysis. By breaking it down to focus on factoring in the categories, the improved response allowed for the additional cross sectional analysis. Saving Changes...
PMI Team Member
