@Nikola Janjic thank you!

I just passed my PMP exam this week, and I used ChatGPT quite a bit along with Claude during my preparation. I ran into exactly what you’re describing. AI was most useful for spotting patterns in my mistakes, not grading answer keys. I uploaded my practice exam results and had it analyze the questions I missed. The night before my exam, I narrowed those down to 14 questions where I most questioned the answers or struggled with the reasoning. Looking at those questions together, we noticed recurring patterns: I sometimes reached for predictive tools in agile situations, focused on internal team processes when the problem was really client satisfaction, or jumped toward a solution before fully understanding what the question was asking. That was far more valuable than arguing over any individual answer. When AI and an answer key disagree, don’t treat it like a court case. Some practice questions are genuinely ambiguous, and two answers can represent different parts of a reasonable sequence. There was one posted by someone else here I was also following where AI and I favored C, but Andrew’s answer was B. Understanding why B was the intended answer, while recognizing why C was appealing, taught me more than trying to prove either one was right. A few things I’d suggest: - Ask AI to make the strongest case for both answers, rather than trying to convince it you’re right. AI can be overly agreeable, especially when you challenge its initial response. - Use AI to identify patterns across multiple missed questions. Are you missing concepts, misreading the question, or selecting a reasonable action that isn’t the best next action? - Keep your course and PMI principles as your foundation. AI is a tutor and thinking partner, not an authoritative answer key. - Don’t spend too long debating ambiguous questions. Understand the principles involved and move on. One of the most helpful habits I developed was asking myself three things before answering: Is this predictive, agile, or hybrid? What exactly is being asked? Who owns the problem?

Awesome, congratulations!

I think @Abdul Moiz is correct to question this. One thing that is stated is before ANY analysis is done. The PM could do some preliminary analysis to bring to the sponsor for providing the change request. Full analysis would be done as a part of the change request. However, where this does stand out - is we are going to the sponsor for the decision. Since this is a change control item, the sponsor MAY be the person making the decision, but we should assume there is a change control board. The sponsor may be a part of that, but we should not assume that because they are proposing the change, that they are the one making the decision. The decision needs to go through the proper process. @Claire Lai, a note I would make is that the C answer is bringing the analysis (the work) to the sponsor for THEIR decision, not asking the sponsor to do the work of the analysis. Small clarification. One could argue the PM submits the formal change request after a sponsor, team member, customer, etc. identifies a need for the change. So I would say neither answer is actually correct, but B is the -best- answer because what should be done *next* is document the request, then analyze, then submit for review to the CCB/or designated change authority.

@Claire Sung I was in the same boat - then the ultra hard questions got me feeling less certain. I then just ended up taking the questions I got wrong and revisited them again. There were only a handful where I still thought there was confusion around the question / answer options. But after reviewing several of them, following the PMP mindset more closely I would have selected the correct answer. So my takeaway was that the ultra-hard are really there to push the limits of looking at really close answers - but because the video was so different, I found that less helpful. I felt a lot better after just reviewing the incorrect responses and comparing my logic. Then reviewing if I was misaligned in how I was following the mindset. Good luck!