Director, Learning Design & Development| PMIAsheville, NC, United States
Validating and checking outputs is critical when working with AI systems like Generative AI. Such validation approaches may include establishing clear criteria, implementing strong testing protocols, and continuous refinement.
In your experience with AI, what are some best practices for ensuring the results you receive are accurate, relevant, and aligned with your original goals?
I define my kpi's when prompting this also helps validate the output Saving Changes...
Sherko IsmaelBPO Director| Daw Al FadaSulaymaniyah, BG, Iraq
Be specific, determine the dimensions, set the goals, and describe the output format.
it is also important to sequence the requests to AI to do it and not ask multiple requests together, this to avoid a confusion and missing AI response to some of the requirements. Saving Changes...
I have almost the same approach to the matter as the most of you. This is my list:
1. Break the question you are going to ask AI parts, which could be solved separately or in chain; try to avoid general questions.
2. Be precise and clear.
3. Avoid jargon and specialized terminology or explain it if you definitely need to use it.
4. Specify the context for all of your requests, upload documentation to base the output and use examples
5. Provide AI with a desired outcome and format of the answer you expect to get.
Try again if you are not satisfied with the result.
Hello Jabin, Completely agree that it is very similar to mentoring associates! The ability to take a defined character or role and tailor the approach or answer to a specific question. Most of the time you are, as the mentor, peeling back the layers to get to the actual problem or question all while re-evaluating responses based on output received from the mentee. Saving Changes...
Getting accurate, relevant, and goal-aligned results from AI systems depends largely on how you interact with them, it depends how accurate data you provide and how full details objectives as well. Saving Changes...
Pablo PosliguaGerencia de Proyectos| IndependienteQuito, P, Ecuador
Compare the results with the initial information and define a learning period and adjustments so we can trust the data. Saving Changes...
A very good question and also difficult to answer as well. However you have to go to the basics and say as far as you are concerned, how well are you versed with the subject at hand ?. There are facts which the AI will generate and if you can verify these facts the more reliable the generated response will be. The fewer the facts then it means that the Generative AI response is far from meeting your original goals. Then it becomes very critical that you review the accuracy , relevancy and the alignment of the response to your original need. Unfortunately there are no clearly defined metrics that one can use a model to evaluate an AI generated response. So from my personal experience I basically restrict AI to an area where i have sound knowledge of , else it becomes almost impossible to verify details generated by an AI if you venture into unchartered territory. However with long usage and exposure your confidence also tend to increase as well. The best practice and protocol to follow would be to consult subject matter expects to validate the AI generated response before making critical decisions based on it to avoid any inherent associated risks which you might be not aware of.
I normally go in discussion mode with AI to refine the responses I get. Somehow I use the RTF and CREATE model to get responses that are relevant and to the point. Saving Changes...
Be specific and clear on the query. Validate the results with other AI to ensure its correctness.
Quote better examples to AI to understand and allow AI to ask queries to ensure the understanding on the query is right. Just like how we ask to our team member. Saving Changes...
By ensuring your prompts are very clear and detailed, providing examples or reference materials where possible and requesting validation of the AI provided statistics to avoid hallucinations Saving Changes...
As I am learning more about AI, I realize I need to consider AI as a consultant or a team member on a project. This means, prompts need to define the objectives of the project clearly, provide the necessary data for assessment/evaluation/ decision making, and set the right expectation for the outcome. Just like communicating with a human, clarity, quantity and expectations need to be set, even if it requires iterative attempts. Saving Changes...
PMI Team Member
