If you’re a people person, enjoy organizing, or prefer talking things through and working in a group, use this Local Facilitation Guide to host a micro-consultation! This can be done at your home, school, workplace, public library, studio, or wherever else you gather with people. Bring together your friends, family, students, coworkers, teammates, bandmates, or classmates. Work through this Guide together, and designate someone to take notes as you discuss. At the end, decide as a group what your answers are to questions of your choice in the Submission Template and complete it together for sending into the consultation.
NOTE: Submissions will be posted publicly, so please only share details you are comfortable being posted. Keep this in mind if sharing stories or experiences that are not your own, and do not share identifying details about someone without their consent.
Preparation
Participants: 3-8 people (recommended, but feel free to invite as many people as you’re comfortable organizing)
Time: 2 hours
Recommended Materials:
- Markers and Pens
- Sticky Notes
- Notepads
- Place to put sticky notes (a table or wall is fine, but you can also use a whiteboard or blackboard)
Pre-Reading:
- Go through the list of bulletpoints in the Open Letter for examples of AI in the news
- For more in-depth learning, you may select something from the Resources List to work through as a group.
- Be prepared to talk about your experiences with the many things called AI
The goal of the facilitation is to come to consensus as a group about possible AI futures to help draft a letter for the Public AI Consultation.
Activity #1: AI Icebreaker (20 minutes)
Goal: Introduce the group and their attitudes about AI to each other.
Process: In this activity, participants share and explore the impacts of their personal experiences using, encountering, or being affected by AI in their lives and communities.
Please break into pairs, and be prepared to interview each other. Each two-person group will take turns being interviewer and interviewee. The interview should ask:
- What is your name?
- What has been your experience with AI?
- Why did you attend this event?
At the end of the event, each interviewer should introduce their partner to the group. Everyone then goes around being an interviewer to introduce their partner and an interviewee where they are introduced by their partner.
Activity #2: Our AI Future? (40 minutes)
Goal: Determine different AI futures that participants might want to encourage or oppose. Participants are responsible for imagining what their futures, and their community’s and society’s future, might look like under different circumstances involving AI, by extrapolating from present-day headlines.
Process: They will do so by reflecting on news headlines and critically interrogating the values that guide different futures from the reading lists on the site.
Step 1 collects signals from participants:
- Participants selects 1-2 stories from the Inspiration lists. Remember stories do not all need to be about AI especially for positive stories.
- On sticky notes, ask people to describe their stories. Spend a minute writing a description of your story and its policy links if included in the reading list.
Step 2 aims to create a messy map of probable futures. Ask people to share their stories and post their sticky notes on a shared wall. They should post their stickies in the center of the wall. Then have participants inspect the wall and the stories shared.
The goal of Step 3 is to divide stories into 2 futures, a desirable future and an undesirable future. Some stories might be in the middle or not quite fit and that’s fine. and, if as a group try to sort the stories into three clusters.
- Select a reporter (someone who will report back to the large group) and someone who will scribe (e.g., write the themes) for the next Activity;
- Select a facilitator who will move stories into one of the two futures.
- Group should try to come to consensus about whether each story is a signal for a desirable future or an undesirable future, and why
- Identify (as a group) what themes and how would you describe each future.
Activity #3 Taking Action (40 minutes)
Goal: Draft a letter for the People’s Consultation based on the futures from Activity #2.
Process: Have the Reporter from Activity #2 report back. Review the questions from the submission template. Break into groups with a key writer to use the Reporters notes to draft responses for key sections.
Closing (20 minutes)
Groups report back and read their sections. Assign someone to tidy up your notes for a Draft response to the Public Survey. Set a deadline for submitting the materials to the People’s Consultation
_________________
Optional Activity: Bad Bots – Exploring LLMs
The goal of this activity is to introduce participants to how LLMS think. This activity is based on a workshop model developed by a group of students at SFU named Callum Barker, Mahin Khanna, Shendian Zeng and Yeqi Sun.
Objectives:
- Illustrate how LLMs reproduce or amplify human errors and biases;
- Highlight the fallibility of LLMs;
- Demonstrate how easy it is to be misled by these tools;
- Demonstrate the importance of protective, supportive or educational interventions.
Required Resources:
You’ll need several laptops running LLMs. The LLMs need to be prepared ahead of time.
Option 1: Pick a single LLM (such as ChatGPT). Before the activity starts, give the LLM on each machine a different set of instructions about how to answer questions. For example, tell one instance to answer all questions as conservatively as possible, tell another to answer all questions critically, and tell yet another to offer neutral answers. Or tell one instance to answer questions using only information from Wikipedia, and another to answer questions using only data from Reddit.
Option 2: Pick three different LLMs (such as: ChatGPT, Gemini, Claude, DeepSeek, LLaMA, Mistral, Command R, Falcon, yi, Orca, etc.) and run a different one of each laptop.
Here’s some resources to help understand how this exercise might work:
- https://github.com/vectara/hallucination-leaderboard
- https://www.linkedin.com/pulse/how-llms-think-differently-elina-halonen-mwnje/
Activity
Break participants into groups. Don’t tell them anything about the LLM they have been given. Give them a list of questions to ask the LLM.
Questions or prompts could include things like:
- Summarize the top 3 major world events of this year.
- What is the best way to study for a test?
- How should I discipline my child when they misbehave?
- Summarize Carney’s major policy initiatives on X topic.
Come back together and discuss the similarities and differences between the answers provided by the various LLMs.
Reflection
Spend some time considering what emerged through this experiment:
- What did you learn about how LLM’s work from this experience?
- What are you not able to know about LLM’s? What would you like to be able to know?
- What are the opportunities and what are the risks or dangers of relying on these tools?
- How might these tools shape our labour, environmental, cultural, identitary, collective, etc. experiences. (See examples of stories.)
- What protections, supports, infrastructure or interventions are required on the basis of this experiment?