Scrum.org AIF I: AI Fluency for Product Team Members I (AIF I)

Last checked November 10, 2026 against the official Scrum.org exam page.

1. Question 1AI Foundations

   A product manager asks a language model to summarise a customer interview transcript. The summary includes a direct quote attributed to a customer, but that quote does not appear anywhere in the transcript. What best explains this?

   1. AThe model searched the live web for a matching quote and returned a cached copy of it
   2. BThe model generated text that is statistically plausible but not grounded in the supplied transcript, which is a hallucination
   3. CThe model stored the quote from an earlier customer conversation in its long-term memory
   4. DThe model matched the interview against a fixed template library and filled in a stored quote
   Show answer and explanationHide answer

   Correct answer: B

   A hallucination is output that reads as plausible but is not supported by the input or by reliable knowledge, and the quote here has no source in the transcript. The summary was not produced by a web search, a memory of another customer, or a fixed template, so the fix is to ground the summary in the transcript and check every quote against it.

2. Question 2AI Prompting & Iteration

   A product manager types 'write user stories for our app' into an AI assistant and gets generic stories that do not fit the product. Which change is most likely to improve the result?

   1. AAdd the product context, the target users, and the format and constraints the stories should follow
   2. BRepeat the same vague request several times in a row and keep whichever response happens to be the longest of them
   3. CAsk for the answer in a single word so that the model is forced to be concise
   4. DStart a new chat and send the same request again with no other details added
   Show answer and explanationHide answer

   Correct answer: A

   Specific context, audience, and format give the model what it needs to produce output that fits the product, which is the core of effective prompting. Repeating a vague request or shortening the answer does not add the missing information, and a new chat with the same vague request will tend to return the same generic result.

3. Question 3AI Quality & Confidence

   A product owner has an AI assistant draft acceptance criteria for a checkout feature. Before sharing them with the team, which check best supports confidence in the draft?

   1. AConfirm that the draft reads as finished, with detailed and confident wording throughout
   2. BCheck each criterion against the source requirements and against what the team knows about the product
   3. CShare the draft with stakeholders immediately, since fluent and well-structured language is a reliable sign that it is correct
   4. DRun the same prompt once more and accept whichever version appears first on screen
   Show answer and explanationHide answer

   Correct answer: B

   Acceptance criteria must match the real requirements, and only a comparison against those sources and the team's product knowledge can confirm that. Fluent wording, a finished-looking layout, and the first result on screen are not evidence that the content is correct.

4. Question 4AI Accountability

   A team uses AI to draft pricing emails for customers. One email goes out with an incorrect discount. Who is ultimately accountable for that message?

   1. AThe AI tool vendor, since the tool generated the wording that was then sent to the customer without any edits
   2. BNo one, because AI-generated content is not owned by any person or organisation
   3. CThe team and organisation that chose to send the message, with a named person who approved it
   4. DThe customer, who should have checked the discount before replying to the email
   Show answer and explanationHide answer

   Correct answer: C

   Organisations and the people who approve AI output remain accountable for what they send, whatever tool drafted it. The vendor, nobody, and the customer are all wrong because accountability does not transfer to a tool or to the recipient.

5. Question 5AI Foundations

   What is a token in the context of large language models?

   1. AA signed certificate that proves a user is allowed to access the model's API
   2. BA saved conversation thread that the model can reopen in a later session
   3. CA numeric rating that the provider assigns to each user based on how often they send prompts and how long those prompts are
   4. DA unit of text, often a whole word or part of a word, that the model reads and produces
   Show answer and explanationHide answer

   Correct answer: D

   Models read and produce text as tokens, which are chunks such as whole words or word fragments, and context limits and usage are measured in tokens. The other options describe access credentials, saved chat history, and a provider's rating of users, none of which defines a token.