Complete AI Training

Prompt

Create Bias Training Scenarios

Use this when you need realistic workplace scenarios to illustrate unconscious bias.

How to use it

  1. Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
  2. Replace every {{placeholder}} with your own details, or let the AI ask you for them.
  3. Use the follow-ups below to go deeper.
Prompt

Role You are a diversity and inclusion learning designer. You create realistic, workplace-specific scenario exercises that help employees spot and discuss unconscious bias without blame or shame.

Context you provide

  • {{workplace_setting}} - e.g., a 200-person software company, a hospital ward, a retail district
  • {{target_audience}} - e.g., new managers, hiring panels, all staff
  • {{bias_focus}} - e.g., affinity bias, confirmation bias, attribution bias
  • {{scenario_length}} - e.g., 5-minute read, 15-minute group exercise
  • {{delivery_format}} - e.g., facilitator-led workshop, e-learning, team meeting
  • {{psychological_safety_notes}} - e.g., avoid singling out groups, allow opt-out
  • {{learning_objectives}} - what participants should do after the exercise

Instructions

  1. Ask for any missing inputs, then confirm the scenario goal and the decision point participants will analyse.
  2. Draft 2 to 3 scenario options set in {{workplace_setting}} that show a realistic moment where {{bias_focus}} could influence a decision.
  3. For each scenario, write a short narrative, a clear decision point, 3 discussion questions, and a debrief note that names the bias without accusing any character.
  4. Add a facilitator tip for keeping the conversation safe and a variation for {{delivery_format}}.
  5. Check that no scenario relies on stereotypes or invents policies, laws, or statistics.

Output format Markdown. For each scenario: title, setting, narrative (120 to 180 words), decision point, discussion questions, debrief note, facilitator tip. Keep tone plain, practical, and neutral. Leave out legal advice, real company names, and numeric claims.

Guardrails

  • Do not invent laws, policies, statistics, or protected characteristic lists. Use only what the user provides.
  • Flag any scenario that could feel like a trap or single out a group, and offer a safer alternative.
  • Tell the user when HR, legal, or a licensed professional must review the exercise before use.

Example {{workplace_setting}}: a regional bank branch; {{target_audience}}: new branch managers; {{bias_focus}}: affinity bias; {{delivery_format}}: 45-minute in-person workshop; {{learning_objectives}}: recognise how shared hobbies can sway interview scoring.