Professional learningfor Teachers, Librarians, Coaches, Faculty
The Human Language Model
Educators become a tiny language model with paper slips and a cup, and discover firsthand why fluent AI text can be confidently wrong.
All activities Professional learning
Hype Check Showcase
A tech showcase remix: teachers rotate through emerging-tech stations, including AI and synthetic media, and have to separate the vendor's claim from the learning evidence before they give a verdict.
Overview
Traditional tech showcases are tool petting zoos: people leave excited about a gadget and unsure what students would do with it. This remix keeps the energy and adds a hype check at every station. Groups read a claim, try the technology (or examine a printed stand-in), and must record what learning evidence it could produce, what could go wrong, and who might be left out. Two stations involve synthetic media, so teachers practice the detection-and-context moves their students need.
Objectives
- Participants will distinguish a technology's marketing claim from the evidence that it improves learning.
- Participants will identify limitations, bias, or misuse risks in at least two AI or synthetic media tools.
- Participants will give each station a Pilot, Watch, or Pass verdict and justify it with a specific classroom use and a piece of student evidence.
Materials
On paper
- Handout A: Station Cards (print one card per station, large, and tape it to the table)
- Handout B: Hype Check Log (1 per person)
- Handout C: Showcase Remix Planner (1 per coach or team)
- Chart paper titled "Room verdicts" with columns Pilot, Watch, Pass; dot stickers in three colors
On screen
- One device per station running the technology (district-approved tools or accounts only)
- A short synthetic voice clip made from the facilitator's own voice, with the facilitator's consent, for Station 1
- A projector for the opening demo
Before you start
- Choose 5–7 stations from Handout A that match what your district can legally and safely demo. Swap in real tools you already have approved; keep the claim and questions on each card.
- Set up each device logged in with a facilitator or demo account. No participant should need a personal account on a consumer AI tool.
- For Station 1, make a 15-second synthetic clip of your own voice saying a harmless line and record the same line yourself. Never clone anyone else's voice.
- For any station without a working device, print the card's back text as the stand-in artifact so the station still runs.
Step by step
-
10–6 min
Hook
Which one is really me?
Play the two clips of you saying the same sentence: one recorded, one synthetic. Everyone votes. Reveal. Ask: "If a student got a voicemail in the principal's voice asking them to share a password, what would you want them to do?" Say: "Today every exciting tool comes with a hype check. The question isn't 'Is it cool?' It's 'What evidence would show it helps students learn, and what could go wrong?'"
Facilitator noteMany rooms guess wrong or split. The honest takeaway is that listening harder is not a reliable detection strategy. Checking the context and the source is.
-
26–13 min
Model
One hype check, out loud
Model the four questions on the Handout B header using Station 3 (the AI writing detector): Claim: "Detects AI writing with near-perfect accuracy." Who says so? The vendor. Learning: What would students do differently? Nothing: it is a policing tool. Evidence: What would show it works? Independent testing on your own students' writing, including multilingual writers. Risk: A false accusation lands on a real student. Verdict: Pass, or Watch at most, and never as the only evidence.
Facilitator noteDetectors make a strong model because the gap between claim and classroom reality is easy to see. Don't claim a specific accuracy figure; the point is that a vendor's own figure is a claim to verify.
-
313–48 min
Explore
Station rotation
Groups rotate every seven minutes through five stations. At each: read the claim, do the "Try it" task on the card, then complete one row of Handout B. Before leaving, each group places a dot on the station's card for Pilot, Watch, or Pass. Groups may not give a Pilot verdict unless they can name one student task and one piece of evidence the task would produce.
Facilitator noteCirculate with one question: "What would a student turn in that shows the learning?" If a group can't answer, the verdict should be Watch.
-
448–58 min
Debrief
Room verdicts
Tally the dots onto the "Room verdicts" chart. Ask the group with the most confident Pilot to make its case in 60 seconds, and the group with the strongest Pass to respond. Then ask: "At the two synthetic media stations, what did we learn about teaching students to question what they see and hear?"
Facilitator noteLook for the shift from "Can we spot fakes?" to "How do we check the source and context?" That's the media literacy move that lasts as the technology gets better.
-
558–71 min
Create
Remix your own showcase
Coaches, alone or with campus partners, use Handout C to plan their own campus showcase with a hype check built in. They pick three stations their teachers are asking about, write the claim and the "Try it" task for each, and decide what they'll collect afterward to see whether any teacher actually changed practice.
Facilitator noteEncourage one station that is unplugged or low-tech. It signals that the showcase is about learning, not screens.
-
671–75 min
Transfer
One follow-up
Each person names one station they'd explore further with a single class and writes the student evidence they'd bring back to a coach or PLC in two weeks.
Facilitator noteCollect Handout C from coaches if you want to support their showcases.
Paper or screen
Unplugged
Each station card's back doubles as a printed stand-in: a transcript of the synthetic voicemail, a written description of what the image generator produced, a sample of the AI tutor's dialogue, the dashboard's printed risk list. Groups run the same hype check on those artifacts. It works surprisingly well, because the four questions are about claims and evidence, not the experience of the device.
Digital
Run each station on a device using district-approved tools with facilitator accounts. Groups log their rows in a shared form so the room's verdicts update live on the projector. For a virtual showcase, each breakout room hosts one station with a screen-shared demo and groups rotate rooms.
Does it need a screen? Trying the tools is what shows teachers how convincing a synthetic voice is or how quickly an image generator reproduces stereotypes; paper descriptions can't match that. The hype check keeps the hands-on time focused on evidence rather than novelty.
Evidence of learning
What you should be able to see or collect if it worked.
- Every Pilot verdict on Handout B names a student task and the evidence it would produce.
- At the synthetic media stations, groups describe checking source and context rather than relying on spotting flaws.
- Groups name at least one limitation or bias in an AI tool (for example, stereotyped images or confident wrong answers).
- Remix plans on Handout C include a follow-up step to see whether any teacher changed practice.
Adaptations
- Elementary campuses
- Keep Stations 1, 2, and 5. Frame the questions for young students' safety: "How would a 2nd grader know this isn't a real person?"
- Higher Ed
- Replace the smart glasses station with an AI research assistant that summarizes papers. Trace its citations to see whether they exist and say what it claims.
- Short version (30 minutes)
- Run three stations at five minutes each with no Create step. Coaches take Handout C home.
Standards connections
Teachers test tool claims and synthetic media at stations, preparing students to choose tools for learning and check media by source and context.
- TEKS
- practical technology conceptsethics and laws (c)(9)TEKS sections: Technology Applications §126.5–§126.7, §126.8–§126.10, §126.17–§126.19, high school Technology Applications courses (19 TAC Chapter 126)
Reflect
- In what ways am I helping teachers meaningfully integrate technology into their teaching? (Coach ELE guiding question)
- How do I stay current on emerging technology without passing the hype along to teachers?
- What would convince me, six weeks after a showcase, that it changed anything in classrooms?
Take it to your students
Teachers try one Pilot-verdict tool with one class and bring the student evidence to a PLC. Coaches run a campus showcase from Handout C and follow up two weeks later with one question: "What did students produce?"
Pairs well with
- Critical Thinking Breakouts: Signal Check, The Viral ClaimA student follow-up on checking viral and manipulated media (grades 6–8).
- Gen AI Literacy Breakouts (K–8)Student activities on how AI tools work and where they go wrong.
- Augmented K–16 AI Learning FrameworkA framework for deciding where AI belongs in a learning task.
