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?
When I´m trying a new topic on the AI, I start with the general task and then , iteratively, improving the prompts by been specific, changing the way it shows me the output or going deeper into the question, including new information in each prompt depending on the previous result. Saving Changes...
Always critically evaluate the AI's output. Cross-reference information with reliable sources, especially for factual content. Saving Changes...
Piotr KarwotProject Manager| DXC.TechnologyCieszyn, Poland
Provide context, example, be clear and specific, use iteration, break down complex tasks to small ones, use role base prompts, do not forget to protect sensitive data and finally... understand that AI is a tool with some strengths and limits :) Saving Changes...
The AI implementation should remain relevant, accurate and aligned with changing conditions over time. This approach involves regularly retraining models with new data to prevent performance degradation, as well as monitoring model outcomes to detect any biases or inaccuracies that might develop. Saving Changes...
Giving feedback to LLMs also helps improving next result. I do this often times when the prompt yields the desired result. Simple word such as Good! helps create context for the next prompt. In most cases after recursively refining the prompt, LLMs can give solutions such as final fix and truly the final fix works Saving Changes...
In my experience, some of the best practices for ensuring accurate and relevant results from AI include -
1. Be clear and concise
2. Provide a context
3. Define the persona
4. Define the result that you are expecting
5. Refine the responses basis your need
6. Cite the source
7. Provide examples of what you require Saving Changes...
Robert FourieProject Management| ArxadaMargate, South Africa
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? Saving Changes...
Luis Carlos BejaranoProject Management Specialist| Centro de Tratamiento e Investigación sobre el Cáncer - Fundación CTICBogotá, Colombia
1. Clearly defining objectives before interacting with the AI.
2. Designing specific and contextual prompts, avoiding ambiguity.
3. Validating results by cross-checking with reliable sources or previous data.
4. Iterating and refining prompts based on the outcomes received.
5. Involving human experts when critical thinking or strategic decisions are required.
These practices help align AI use with project goals and ensure the quality of deliverables.
Based on my own experience, I’ve found that being clear about my goals and giving detailed context helps AI give me relevant answers. I always double-check important information with reliable sources. I refine my prompts as needed so the results align with what I’m looking for. I keep in mind the AI’s limitations and ask for explanations when I need them. I’m also careful not to include sensitive information in my requests — after all, I’d prefer some details don’t end up in some AI’s stand-up routine! :-) Saving Changes...
Previously, before knowing how the CREATE formula can be utilised, I'd ask specific questions - without being personal. The responses would be specific to the question but not directly contextual. I've learned that providing context is key...refining is inevitable, but within the context provided.
The key is to simplify work, not avoid it altogether. Saving Changes...
PMI Team Member
