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?
AI is a broader term. Generative AI is just an ancient model but everything "explode" when Google published the new architecture called transformer in 2017. So, with that said, take into account that generative AI is just "predictive test with steroids" just simplifying the model. With that said, two key points has to be taking into account when somebody works with AI: 1-human in the loop. 2-AI without Data (today called data science discipline or big data or whatever) is the same thing that live without oxygen. Talking about generative AI all related to technology has almost not impact with relation to all related to non-technological roles and activities. What you stated about accuracy and things like that are easy to implement because there are a lot inside disciplines like statistics. Most of them to make things "a priori" to prevent instead of cure. Few organizations taking into account that when generative AI environments are put in place almost a new business unit has to be created where roles like lawyers, linguistic, diversity and inclusion specialist must be hire to help on put it in place.
Be specific, provide examples, use a persona. Saving Changes...
Like with any new tool, you need to test the results before you scale up.
Think about if you were to manually model a very complex problem in a spreadsheet. You don't build all the links and formulas first and then evaluate your final output. You build and test sections of the bigger solution first and then add on layers once you have validated the functionality.
I completely agree with this response, you need to test the results as you receive them, but also validate the information with other sources. Saving Changes...
I particularly like the tools and examples available in the PMI course for Prompt Engineering for Project Managers on how to use the CREATE formula prompt for tailored response output. Saving Changes...
Some of my items may be redundant but the most important things in my experience so far is:
Be precise and clear.
Be sure you explain jargon or specialized terminology
Provide the context for all of your requests
Be sure you provide the outcomes you are expecting
Experiment and refine as you go
I've found breaking down big problems can be better refined by chunking the whole into natural sections and working to refine each section and then working to put them back together.
good point on providing the outcome. Saving Changes...
Jim FuhringPMP | CSM | ITIL | RUP | Technology and IT Senior Leader| The Trade DeskOxnard, Ca, United States
To prompt the AI to provide accurate, relevant, and output aligned to my goals I use prompt formulas. For the CREATE method formula I specify the final "E" Evaluation Criteria in the following ways...it is effective to use combinations of the below...also, ask the AI to provide the sources for where the information came from. It will provide you hyperlinks that you can verify
Completeness:
"The response must cover all relevant aspects of the topic in detail."
"Ensure that the answer addresses each part of the question."
"Provide a comprehensive explanation, including both the pros and cons."
Clarity and Readability:
"The explanation should be easy to understand, even for someone without technical knowledge."
"Use clear and simple language without jargon."
"The response should be written at a reading level suitable for a general audience."
Brevity or Conciseness:
"Provide a concise summary of the key points in no more than three sentences."
"Keep the response under 200 words."
"Summarize the main ideas, avoiding unnecessary details."
Depth of Analysis:
"Provide an in-depth analysis of the topic, including relevant historical context."
"Offer a detailed breakdown, going beyond surface-level explanations."
"Include advanced technical insights to cater to an expert audience."
Formatting and Structure:
"Provide the response in bullet points for easy readability."
"Ensure the answer is structured into clear, logical sections."
"Use a table format to compare the key features of each option."
Creativity and Originality:
"Propose unique, out-of-the-box solutions to the problem."
"Offer innovative ideas or perspectives that haven’t been widely discussed."
"Generate original content rather than relying on common examples."
Timeliness:
"Provide the most current information as of September 2024."
"Ensure the data reflects the latest advancements in the field."
"Include any relevant updates or changes from the past year."
Relevance:
"Ensure that the response is directly related to the question."
"Only include information pertinent to the specific scenario."
"Avoid extraneous details and stick to the key issues outlined in the query."
Actionability:
"Provide actionable steps that can be implemented immediately."
"Ensure the recommendations are practical and feasible."
"Include suggestions that are directly applicable to real-world situations." Saving Changes...
Manuel OspinaDirector PMO| Expert Projects S.A.S.Santiago De Cali, Valle Del Cauca, Colombia
The use of AI to define and refine these objectives can be useful, as it can analyze large volumes of historical data and provide projections or recommendations.
AI tools can help analyze the scope of changes in requirements and their impact on timelines, costs and resources.
AI-based tools can predict potential deviations or inconsistencies, enabling early intervention.
AI can analyze performance patterns and predict potential problems before they materialize, enabling real-time adjustments. Saving Changes...
In order to get an accurate output there will be required to use "Create" model in prompts and apply agile iterative approach to deal with uncertainties by breaking down risks. Saving Changes...
Give a clear and specified prompt full of details, analyze and reiterate the prompt to sharpen the response.
As if you're cooking a meal you keep trying to enhance each ingredient and part of the meal by adding more suitable flavors. Saving Changes...
Helena GenerPM I| Setting Consultoria en T.I., S.L.Barcelona, Barcelona, Spain
Hi,
In my experience provide an detail context of my request, use an specialiced terminology, be precise and clear, provide the outcome expected and iterate and iterate if its necessary Saving Changes...
To ensure AI results are accurate, relevant, and aligned with your goals, clearly define your objectives and provide specific, detailed prompts. Regularly review and validate the outputs against trusted sources. Continuously refine your prompts based on feedback and observed outcomes.
PMI Team Member
