AI at Work Quiz: Are You Getting the Most From AI Productivity Tools?

A product manager wants AI to identify weaknesses in a business case. Which prompting technique is most effective?

Tell the AI the business case is strong to encourage positive framing Ask the AI to 'improve' the business case without specifying what type of feedback is wanted Persona assignment — 'Act as a skeptical CFO reviewing this business case. Identify the three weakest financial assumptions.' Ask for a step-by-step summary of the document without specifying the feedback goal

A sales team writes a different prompt every time they need to generate a follow-up email after a meeting. What is the highest-leverage improvement?

Create a reusable prompt template with clearly marked variables for meeting date, client name, key discussion points, and next steps Switch to a more powerful AI model Only the most senior team members should use AI — it causes quality variance Each salesperson should develop their own unique prompting style

You use an AI tool to draft a client proposal email. The AI includes a specific statistic — '73% of companies in your industry report cost savings from this approach.' What should you do before sending?

Verify the statistic independently — AI tools frequently fabricate specific data points and percentages Add a disclaimer that the email was AI-generated Remove all statistics since AI cannot know current data Send it — AI models only include verified statistics

A small physical therapy clinic wants to use an AI scheduling chatbot on their website. Patients will input their names, contact info, and appointment reasons (which may reference medical conditions). What is required before deploying the chatbot?

A Business Associate Agreement (BAA) with the chatbot provider, since the chatbot may process Protected Health Information under HIPAA HIPAA only applies to clinics with more than 20 employees Chatbots don't count as HIPAA-covered technology — only EHR systems do A standard terms of service acceptance by the clinic is sufficient

A law student submits a legal brief that was substantially written by an AI, without disclosure, for a law school assignment. The professor discovers it. What is the most likely consequence?

No consequence — there are no rules about AI in academic settings yet The brief is automatically rejected for technical reasons Academic integrity violation — most law schools have explicit AI use policies, and submitting AI work as your own without disclosure violates them The student receives partial credit since the AI wrote it, not the student

Which AI tool is purpose-built to automatically transcribe and summarize meetings, extracting action items?

ChatGPT — by manually typing what was said in the meeting Adobe Acrobat AI — designed for PDF document review Meeting transcription tools like Otter.ai, Fireflies.ai, or Microsoft Copilot in Teams Perplexity AI — which is designed for web research

Which prompting technique best helps a professional use AI to identify blind spots in their own analysis?

Ask AI to score the analysis out of 10 Ask AI to write a better version of the analysis Ask AI to summarize the analysis back in its own words Ask AI to list the critical assumptions the analysis relies on, then ask what would happen if those assumptions were wrong

Which of these is the most effective prompt for generating a useful competitive analysis?

'You are a market analyst. Compare the pricing, integrations, and enterprise support of ProductA and ProductB in a two-column table. Target audience: VP of Sales.' 'What do you think about these two products?' 'Tell me about ProductA and ProductB' 'Compare these products somehow'

A company uses AI to screen resumes and notice that the tool systematically ranks candidates from certain universities higher. What is the most likely cause?

The AI was trained on historical hiring data where those universities were over-represented among successful hires — encoding past selection bias The university ranking reflects the AI's access to university prestige databases There is no cause — AI resume screening is required to be unbiased by design The AI is correctly identifying the objectively better universities

Which of these professional tasks is LEAST reliably handled by current AI tools?

Summarizing a long email thread Generating a first draft of a project status update Making final strategic judgment calls that require understanding organizational politics and relationships Translating a document into another language

Which of these tasks represents the BEST use of human professional judgment rather than AI?

Generating three options for a marketing tagline Translating a document to another language Deciding how to handle a sensitive interpersonal conflict on your team based on the history and relationships involved Drafting a meeting follow-up summary

A small business wants to use AI for drafting responses to customer complaints that include customer names and order details. What is the simplest way to reduce privacy risk while still using AI?

Anonymize the data — replace the customer name with 'Customer' and describe the order generically — before sending to the AI tool Never use AI for any customer-related communications Ask the AI not to store the customer's name after the session Use AI only on Tuesdays when server load is lower and data is more secure

An AI customer service chatbot performs measurably worse for customers who use non-standard English syntax. This is an example of which type of AI bias?

Training data bias caused by outdated information Performance disparity bias — the model performs less accurately on language patterns underrepresented in its training data Recency bias from the model's training cutoff date Representation bias in AI-generated images

A marketing team asks AI to generate 10 example customer personas for their advertising campaign. After reviewing the output, all 10 personas have similar demographic characteristics. What should the team do?

Accept the default output — AI generates representative samples of actual customer data Replace AI with manual persona creation to avoid bias entirely Add a disclaimer to the campaign noting AI was used Explicitly prompt the AI to generate diverse personas across age, gender, background, and geography to ensure more inclusive representation

An employee's organization has not published any formal AI use policy. What is the safest assumption about AI use at work?

Treat sensitive client data, confidential business information, and employee data with the same protection you'd apply to any third-party tool — assume AI APIs are external services No policy means the employee is personally liable for any AI-related incidents No policy means all AI tools are approved for any use No policy means AI cannot be used at work under any circumstances

A financial analyst uses an AI tool to draft a client investment report. The report contains an inaccurate statement. The analyst submits it without review. Who bears professional responsibility?

The analyst — they are accountable for work they submit, regardless of how it was generated The AI tool provider — they provided the drafting capability The client — they requested the report No one — inaccuracies in AI-generated content are treated as errors outside human control

What does 'chain-of-thought prompting' mean?

Instructing the AI to reason through a problem step by step before giving its final answer — which improves accuracy on complex tasks Providing several examples before the main task Asking the AI to repeat your question back before answering Sending multiple messages in sequence to build up a long conversation

A strategy team wants to use AI to stress-test a business plan. Which prompt is most effective for this goal?

'Tell me if this business plan will succeed' 'Act as a skeptical board member. What are the three riskiest assumptions in this plan, and what evidence would change your view?' 'What improvements can you make to this plan?' 'Summarize this business plan in three paragraphs'

A company builds a prompt library in Notion with 30 useful templates. After 3 months, some templates produce worse results than they used to. What is the most likely cause and solution?

The team is using too many templates — reduce to fewer than 10 AI model updates changed behavior — review and update templates when models are upgraded, and record a version history Notion is not a suitable storage tool for prompt templates All templates expire after 3 months due to API policy

What is the most effective first step when asking an AI to write a professional email on your behalf?

Use the AI's default style so it doesn't sound AI-generated Write the email yourself first, then ask AI to make it longer Provide the context — who the recipient is, what the goal is, and what tone is appropriate — before asking for a draft Ask for the email without context and then copy it directly