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?
Providing the specific context in clear and consise way is essential.
Yes, also ensuring the model has the relevant information first. For closed AI models in particular, it's important to provide the specific documents or other information relevant to your prompt to ensure your output is accurate. You should always validate your responses Saving Changes...
James BradleyProject Manager| Valard ConstructionBurnaby, British Columbia, Canada
To help ensure the results I receive from AI systems are accurate, relevant, and aligned with my original goals, I apply this approach. Starting with the end (output) in mind, I ensure the input is defined and specific and aligned with my original goals. Based on the output, I'll validate the information by referencing third-party (trusted, relevant, timely) source material that AI system used to compile the output, (if relevant) refer to a subject matter expert. Based on validation - I might iterate my prompt to refine the output. I may also use multiple AI tools to cross-reference based on the output I receive and look for patterns in the output (for example, if various AI tools have a similar output for a specific input, my confidence in the output increases). If my confidence isn't increased, I'll iterate my inputs into the AI system(s). Saving Changes...
Anonymous
Ask precise questions to AI as your team member who would help you in building a report. Give feedback both appreciations and improvements. Saving Changes...
Anonymous
I would start by clearly defining the objectives and constraints in the initial prompt in order to ensure AI-generated results are accurate, relevant, and aligned with project goals. Frequently refine inputs based on output quality and project context.
Always validate results using domain knowledge or reliable sources.
Use prompt tracking to maintain consistency across project phases.
Apply human judgment to critical decisions and ensure compliance and data privacy throughout the AI workflow. Saving Changes...
It´s a prompt and then, a think the first attention is accurate the every information.
Be specific, provide context and examples is greatfull to get better results. Saving Changes...
In my experience, one needs to ensure that their requests are clear and precise. If AI's response doesn't meet your expectations then refine it e.g. rephrase your request or ask AI to provide a simpler version of the response Saving Changes...
What are my project goals? who is my audience? what compliance rules do I need to follow? what example documents do I have and what is safe to upload?
2. Provide simple & specific instructions
What do I want my prompt response to look like? i.e. presentation? sprint plan? deep analyses?
3. Kaizen: use iterative prompt refinement
Adjust & refine prompts for better outcomes: reword, revise, provide better input; set a better stage, provide examples, etc. Saving Changes...
David Gonzalez De LaraPROCESS OPTIMIZATION MANAGER| Stevin Rock LLCRas Al Khaimah, United Arab Emirates
The AI is not a mind reader; it's a tool that responds to the information you provide.
Then, before and all the time, we have to define the objective, provide rich context, specify the audience and tone, provide examples, break down complex tasks, and set explicit constraints. Once we receive the output, we must verify critical information, iterate and refine, and check the relevance and bias.
By shifting your mindset from being a passive user to an active AI director, you can consistently generate results that are not only accurate and relevant but also a powerful accelerator for achieving your goals. Saving Changes...
Parke BrownProject Manager| Seed CompanyElizabethtown, Pa, United States
In addition to the great ideas above, selecting a tool you trust enough to share detailed information with is essential. That lack of trust in what you can share with the tool will continually limit how much you sharpen the tool's context. Saving Changes...
PMI Team Member
