Project Management

Please login or join to subscribe to this thread

In your experience with GenAI, how has refining a prompt drastically changed the output quality?

linkedin twitter facebook   Artificial Intelligence  
avatar
Sarah Philbrick
PMI Team Member
Director, Learning Design & Development| PMI Asheville, 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?

Sort By:
< 1 ... 16 17 18 19 20 21 22 23 24 25 26 ... 147 >
avatar
Larry Brod Manager| Baptist Health South Florida Delray Beach, FL, United States
Refining the prompt leads to better outputs, always validate !
avatar
Juan Carlos Mendoza Reto Lima, Lim, Peru
Refining a prompt is very important.
avatar
Daniela Zuppichin Director de Proyectos, Agile Trainer| SiPro360 Project Consulting Ciudad Autonoma de Buenos Aires, Argentina
El refinamiento de un mensaje, ya sea en una conversación casual, una campaña publicitaria o un código de programación, es un arte que puede transformar un simple enunciado en una herramienta poderosa. Un mensaje bien elaborado no solo transmite información de manera clara y concisa, sino que también puede persuadir, inspirar o motivar a otros. Desde la elección de las palabras hasta la estructura de las oraciones, cada detalle cuenta. En la comunicación interpersonal, un mensaje refinado puede evitar malentendidos, construir relaciones sólidas y generar empatía. En el marketing, un mensaje bien diseñado puede aumentar las ventas y fortalecer la imagen de marca. En la programación, un código bien estructurado y documentado es más fácil de mantener y menos propenso a errores. En esencia, el refinamiento de un mensaje es un proceso iterativo que implica la búsqueda constante de la claridad, la concisión, la coherencia y la adaptación al público objetivo. Al dedicar tiempo y esfuerzo a perfeccionar nuestros mensajes, podemos mejorar significativamente nuestra capacidad para comunicarnos, persuadir e influir en los demás.
avatar
Abhishek Gupta Consultant| Teleperformance Delhi, DL, India
In my experience, refining a prompt by making it more specific, structured, or clear can drastically improve output quality. For ex. vague prompts often give generic results, while precise, detailed prompts provide more relevant, and actionable responses. Small adjustments, such as adding context or defining the desired format, does enhance the accuracy and specificity of the AI's output, turning an average response into a highly effective one.
I personally found CREATE to be a great model for prompting. Also, though cumbersome, by prompt results documentation does help in honing your own skills in terms of prompt engineering
avatar
Markus Schuldt Senior Project Manager| ITERGO GmbH Ahrensburg, Germany
A more precise description and detailing of the requirement leads to more precise and better results. I always take a lot of time to create and improve the prompts so that I achieve the best possible result.
avatar
Henry Leal Project Manager| Integrated Project Solutions Georgia, United States
Changing words to elaborate better comprehension and to create specifics results develop a more concise AI answer.
avatar
Oluseun Alamu Arepo, Og, Nigeria
Prompt engineering refinement has produced a better, more accurate and concise response. The more you provide the details and not vague ones the better the results
avatar
Murthy SN Head - Service Excellence, Transformation| Ernst & Young Bangalore, Karnataka, India
Using AI systems depends on what is our purpose for the queries we put in. As an example, let us say there is a content writer who is using it. I would hugely appreciate if the writer creates content first and then uses AI to refine / enhance it after that (instead f the other way). We should not fully depend on AI for the drafts or outcomes. Validation systems and human in the loop is very critical. There could also be certain aspects or exceptions where AI cannot be used or it has not been sufficiently trained). I have used AI for refining official mails (for correcting grammar and sentence formation, formatting), summarizing lessons learnt documents from the project repositories on lessons learnt, summarizing the actions taken report across weeks and few others critical tasks.
avatar
Shelley Matson Vice President of Programs| NextGen RF Design Baltimore, MD, United States
Identifying my audience has greatly increased the readiness of the output; reducing the amount of rework I need to complete
avatar
Michelle Proctor Technology Project Manager| Prince George's County Public Schools (MD) Upper Marlboro, Md, United States
I think utilizing GenAI, is similar in some ways to talking to a human. I think we've all experienced times when even there are misses in our verbal communication with other humans because we each start at a different vantage point, come from a different background, etc. Thus, similarly, there will always be times when we need to restate something or add mor context to receive the type of output/feedback we are seeking. I refine prompts daily and base don't these sessions will now move forward with a more structured approach each time.
< 1 ... 16 17 18 19 20 21 22 23 24 25 26 ... 147 >

Please login or join to reply

Content ID:
ADVERTISEMENTS

"I would have made a good Pope. "

- Richard M. Nixon

ADVERTISEMENT

Sponsors