You are here:

AI-Assisted Audience Intelligence Audit

School of Arts and Social Sciences

About this use case

Students assume the role of a Junior Public Relations Consultant and analyse a dataset of simulated social media conversations. Students first conduct an independent audience before using Gemini or Microsoft Copilot to analyse the same dataset. They then compare both analyses, identify AI strengths and limitations, evaluate AI-generated audience intelligence, and develop evidence-based communication recommendations using professional judgement.

Learning outcomes:

Utilise AI tools for audience analysis and evaluating the effectiveness of public relations initiatives.
Develop ethical frameworks for AI use in Public Relations, considering the challenges and opportunities presented by social media and digital platforms.

The assessment was designed to reflect contemporary Public Relation practice where AI tools increasingly support audience monitoring and sentiment analysis. Rather than rewarding AI-generated answers, students are required to audit outputs, identify limitations and demonstrate professional judgement in communication decision-making.
- Introduce audience analysis and stakeholder mapping.
- Discuss sentiment analysis and social listening
  Demonstrate Gemini and Microsoft Copilot audience analysis capabilities.
- Teach limitations of AI in interpreting sarcasm, stakeholder relationships, cultural communication and misinformation
- Introduce ethical issues including confidentiality, bias, privacy and data sovereignty.
Students are able to:
- analyse audience conversation independently
- evaluate AI-generated audience intelligence
- distinguish between audience sentiment, stakeholder significance, and communication implications
- apply professional judgement to AI-supported communication practice
- reflect critically on ethical uses of AI in Public Relations.

Try it out

This use case requires a medium level of effort to implement.

1. Develop a realistic communication scenario and create a dataset of audience conversations. The dataset should contain a mixture of supportive, critical, sarcastic, neutral, and ambiguous comments to encourage deeper analysis rather than simple sentiment classification.
2. Instruct students to complete an independent audience analysis before using any AI tool. This ensures students develop their own professional interpretation before being exposed to AI-generated insights.
3. Provide guidance on suitable AI tools such as Microsoft Copilot or Google Gemini and require students to analyse the exact same dataset using AI.
4. Ask students to compare their own analysis against the AI-generated analysis. Students should identify agreements, disagreements, omissions, misinterpretations, and differences in prioritisation.
5. Encourage students to test AI outputs through follow-up prompts rather than accepting the first response. For example, students could ask the AI to explain why a theme was prioritised or how sentiment classifications were determined.
6. Require students to provide evidence from the dataset when evaluating the AI output. Students should support conclusions with examples rather than making general statements about AI strengths or weaknesses.
7. Include a strategic recommendation component where students develop communication responses based on their own professional judgement rather than reproducing AI-generated recommendations.
8. Incorporate an ethical reflection section focusing on privacy, confidentiality, data sovereignty, algorithmic bias, misinformation, and practitioner accountability.
9. Require submission of raw AI prompts and outputs in an appendix so that AI use can be documented and critically examined.
1. Emphasise that the goal is not to determine whether AI is "right" or "wrong." The strongest student responses often emerge when students identify nuances, stakeholder implications, or strategic priorities that AI overlooks.
2. Design datasets that include ambiguity, sarcasm, humour, and different stakeholder perspectives. If all comments are straightforward, students have fewer opportunities to demonstrate professional judgement.
3. Assess human judgement rather than AI usage. Students should earn marks for their ability to evaluate, challenge, and contextualise AI outputs rather than for producing sophisticated prompts.
4. Encourage students to examine how AI reaches conclusions. Asking follow-up questions often reveals assumptions, classification methods, or inconsistencies that students can critically evaluate.
5. Use authentic professional scenarios. Audience monitoring, social listening, reputation management, and stakeholder engagement activities closely reflect contemporary communication industry practice.
6. Highlight that low-frequency issues can still be strategically important. Students often learn that audience significance is not always determined by comment volume or sentiment percentages.
7. Build ethical considerations into the assessment from the beginning. Discussions around privacy, AI governance, transparency, and accountability help students understand that responsibility for communication decisions remains with practitioners, not AI systems.
Indirect observation through drafts and process documentation
Students document their assessment process by completing an independent human analysis before using AI on the same dataset. They submit their AI prompts and outputs and critically compare these with their own analysis, demonstrating how their interpretation develops and where human judgement is applied.

Supporting resources

Here are some additional resources that you can browse to help you from TeachHQ:

Learn about AI at your own pace? Enrol in online courses via myDevelopment. Alternatively, you can join ongoing chats about AI and webinars.

Sidebar

Unit: AMU2014	Condition: Individual
Class size: 30–40	Typology: Written
AI approach: Using and applying AI tools effectively	AI statement: Guided use
Visibility of Learning: Indirect observation through drafts and process documentation

About the educator

Validated by

Sri Bala Murugan, Education Designer, Education Excellence

Key tags

Generative AI
AI Literacy
Critical Evaluation of AI
Audience Analysis
Public Relations
Professional Judgement
Ethical AI Use