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?
Aaron GosselinLighting Project Manager| LineWorks EngineeringBrandon, Fl, United States
Booma, I agree with your response! This seems as though it could be quite restrictive in the outputs we receive. As I continue to learn about creating useful prompts I will keep in mind that it may not always be best to use one of the RTF or CREATE to create my prompt and in turn see what kind of creative response I may get from the LLM model. This may change the way I initially input my prompts going forward! Saving Changes...
Vinayak DeoSr. Project Manager| FedExCollierville, Tn, United States
Providing context, specifics about scenario and clearly defined output format have helped me get to the desired results Saving Changes...
George CalouraKey Account Manager - Electric Vehicles| Adler Pelzer GroupRochester Hills, Mi, United States
After reading the comments...I do look forward to using AI in my area of work but am concerned. I understand the more detail you input in the various modules ensures more accurate responses. The concern I have is AI ability to keep up with a lot of us having to make changes on the fly. I agree AI could make our tasks easier and faster, I'm just not confident with AI and adaptability at this point. But I am looking forward to see where this goes. Saving Changes...
There are framewoks to create prompt. This is part of the Prompt Desing discipline. Those that gave me and the initiatives where I was included are:R-T-F (Role-Task-Format), T-A-G (Task, action, goal), B-A-B (Before, after, bridge), C-A-R-E (context, action, result, example), R-I-S-E (role, input, steps, expectations).
Ambiguous wording can confuse the AI. Replacing vague verbs and nouns with precise language makes the desired outcome much clearer. Saving Changes...
Sherif KassemSenior Specialist| Airbus HelicopterAUSTIN, TX, United States
With GenAi, well-crafted prompt can significantly narrow down the scope of possible answers, providing greater control over the style, tone, and format of the generated content, ultimately making the output much more useful and aligned with your needs. Saving Changes...
George Wardell, PMPIT Project Manager| PMI Atlanta ChapterKennesaw, GA, United States
My experience so far with GenAI has been very limited, but I find it as a very useful tool to improve my resume. However, I can see an almost unlimited use case once I am able to land a new job and use my PMP. Saving Changes...
Gerardo IbarraIT Program Manager| Affinity-PetcareBarcelona, Barcelona, Spain
I'm just starting to use GenAI but I realize the importance of being as clear and specific as possible to get better responses ... as we should do when dealing with our colleagues and team members in projects. I guess using these methods will help us all as PMs, to enhance our communication and collaboration in projects and make the best use of this technology. Saving Changes...
Jumarin IsmailDirector Advisory & Forensic Services – Infrastructure & Capital ProjectRiyadh, Saudi Arabia
Great question! In my experience, refining AI prompts can significantly improve output quality. For example, when I asked AI to create a risk assessment framework for audits, my initial prompt "Create a risk assessment framework for audit" was too broad, resulting in a generic and high-level response. Therefore, to get a more useful output, I refined it to generate a structured risk assessment framework for financial audits, including key risk categories, assessment criteria, mitigation strategies, and real-world examples. Thank you.
I achieved the greatest increase in both content and quality of the response through refinement of the persona and the intended audience. Through trial and error I also learned to break the building tasks into palatable segments, each building on the previous output. You can type your initial information, then ask AI to wait for the first task that will include an example response format. At the end of that instruction, again ask AI to wait for the next task. As others have mentioned - dialoging with AI as you would in conversation can continue to refine the output. Saving Changes...
From this course so far, I can tell I've been using AI incorrectly. Watching the examples of how to make AI work in a manner more consistent with what I need will make refining AI prompts much easier as i learn how to "speak" to AI for what i really need. Saving Changes...
PMI Team Member
