Same tools, new federal name. Here’s the vocabulary you’ll see in Idaho and national materials.
AI
Artificial Intelligence
Computer systems that do tasks that usually take human thinking, like writing, summarizing or answering questions. The term Idaho and Playlab use.
≈
SI
Super Intelligence
The federal name since a Sept 29, 2026 executive order. In 2027 national materials, read “SI tools” as the AI tools teams already use.
Presidential SI ChallengeThe national competition for K–12 students and educators. Formerly the Presidential AI Challenge.
Idaho Student AI ChallengeIdaho’s preparation pathway with Playlab and the Idaho Department of Education. Teams can also enter the national challenge if they choose.
Team LeaderThe adult, 21 or older, who registers and leads a team.
AI applicationA tool built on an AI model, with instructions and references, that does one specific job.
Standard TrackAll grade levels. Solve a problem for students, families or communities.
Trailblazer TrackGrades 6–12. A new or unique project that addresses a national challenge.
Educator Category1–3 educators who teach students about SI, or build SI tools that transform school workflows or policies.
First Lady Seal of SI ReadinessSchool‑level recognition for completing SI readiness milestones.
Idaho + national
How they fit
One project, one or two submissions
Idaho comes first, so teams get feedback early. Going on to the national challenge is optional.
Idaho Student AI Challenge only
No requirement to also submit nationally
Submit by Mon, Nov 30
No video required
Virtual feedback from local industry partners, Dec 1–7
Capitol Showcase in January for selected teams
Free trainings, office hours and a Playlab workspace
Title PageProject title, team members with each student’s school and grade, the Team Leader’s school or group, and a video link that works without a password.
SummaryOne page: the goal, who it helps, which SI tools you’ll use, and how the project deepened understanding of responsible use.
The SolutionWhy your idea is the best option, what else has been tried and why it didn’t work, with evidence.
SafetyHow the team protected users’ physical and emotional safety and kept data secure. Worth 10 points.
Learning AchievementsEach member’s role and what they learned, how the Team Leader supported the team, and how you checked SI outputs for accuracy.
Next StepsWhat you’d do with three more weeks, what you’d change for regionals, what it would take to deploy.
Research Supporting the Novel ApproachTrailblazer only. A review of existing tools showing your idea is new.
References & AcknowledgementsFormalizes last year’s citation rule.
Educator details: Impact Potential covers scalability and measurable indicators; SI Accuracy covers an evaluation plan and pilot feedback; Safety covers privacy, bias and accessibility.
Guiding question: How does the Idaho pathway set teams up for the national challenge?
Idaho overview
The Idaho pathway
Support that helps Idaho students succeed.
Students propose AI solutions to real problems in their schools and communities. Idaho teams get training, industry feedback and a chance to showcase their work.
No requirement to go national. Teams can stop at Idaho or use it to prepare for the Presidential SI Challenge.
Teams can use any AI platform. Training attendees also get a private Idaho Playlab workspace.
Oct – Nov
1 · Propose
Teams find a real problem in their school or community and propose an AI solution. No working app required.
Dec 1–7
2 · Get feedback
Every team that submits by Nov 30 gets written feedback from local industry partners, all virtual.
Jan 2027
3 · Showcase
Selected teams present at the Idaho State Capitol during the legislative session, to lawmakers and Idaho employers.
Coming soon
Cash prizes
An announcement for Idaho teams is on the way. Watch your email.
Student work in the building where big decisions are made.
Selected teams bring their submitted projects to the Idaho State Capitol during the legislative session. Playlab staff are on site for last‑mile coaching and tech support.
Present your work
A working project or any kind of presentation.
Answer questions
From lawmakers and Idaho employers.
Get coached on site
Playlab staff help with last‑mile prep and tech support.
Problem Discovery Worksheet. Students list, narrow, and check. Homework: each one talks to a person who has the problem.
2
Decide the AI part
Report back, then answer why this needs AI. Most teams are Track I. Trailblazer is only for ideas that are novel and would work nationwide.
3
Draft
Proposal Template or Trailblazer Plan. Both tracks write a proposal. Cutting to 500 characters is where the thinking happens.
4
Submit together
Tighten, gather uploads, fill in the form as a class. Do not send them home to do this alone.
Four weeks from today is November 30. One period a week fits, and the Educator Guide has the same plan written out with what to watch for in each one.
No video is required for the Idaho submission. Teams that continue to the Presidential SI Challenge will need a single video, 4 minutes maximum, with the national submission.
Idaho overview
What is in it
The template is the form.
Every box in the template matches a field on the submission form, in the same order, with the character limit beside it. Teams draft in the doc and paste across.
Why it helps: 500 characters is about 80 words. Students write long first, then cut. The cutting is where they work out what they actually mean.
Track II teams use the Trailblazer Plan, which covers these same four fields plus the extra evidence a novel, nationwide idea needs.
Write it last. Who, what problem, what the tool does.
500 characters
Problem, question, or need
The one that decides the submission. Most time here.
500 characters
Student work vs AI work
Be specific. Honesty reads as strength.
Optional
Comments for reviewers
Constraints, what failed, what is next.
Part 1 · Next steps
Part 1 wrap‑up
So, what’s next?
Registration is a two‑part process: Idaho first, then national. Start early.
1
Check your Team Leader eligibilityNational21+, and a full‑time teacher, approved club leader or NSF PI
2
Collect proof and consent formsVariesWhat’s required depends on which challenge you’re submitting to
3
Get your team startedBothShare the AI Impact Challenge Student Course
4
Outline with the required sectionsBothPlan the Safety section early
5
Submit to Idaho for feedbackIdahoMon, Nov 30
6
Register and submit nationallyOptionalDec 18 · Dec 23, 3pm MT
Part 1 · Questions
Questions?
Coming up next
Part 2: Building with Playlab
Explore community apps, learn how to build one and set up your team’s space. Stay if you can. If you need to head out, you have what you need to get your team started.
Email rebekha@playlab.ai
Part 2 · Playlab
Stretch
Five minute break.
Grab water. Stand up. We start again right on time.
Staying for part two?
Two small asks, and they make the next hour much better.
1
Turn your camera on
It is a workshop, not a webinar. I would rather see you than a grid of initials.
2
Build along with me
Open a second window and actually make the thing. You will leave with an app instead of notes about one.
Part 2
Playlab’s purpose
Guiding question: What makes Playlab unique?
Playlab’s purpose
What we believe
Breakthroughs in education won’t come from us. They’ll come from the 50 million educators and 14 million high school students who, given the right tools and conditions, are best placed to figure out what AI can do for them and their communities.
Playlab’s purpose
Where educators shape AI
A nonprofit dedicated to creating a future where educators and students around the world have the agency to not just use AI, but to shape it.
CreateShareEvaluate
Playlab’s purpose
Most AI tools hand you a template.
Playlab hands you a toolbox.
01
Build the tool. Any model, your context.
02
Analyze usage. See every conversation, with safety flags.
03
Connect your curriculum. Grounded in your content and standards.
Playlab’s purpose
AI agency is a core competency of the future.
AI agency
AI literacyUnderstanding how AI works, its benefits and its limitations.
Computational fluencyThe ability to design and build systems with technology.
DiscernmentThe ability to evaluate AI systems for outcome and impact.
Playlab’s purpose
Responsible AI is built on a foundation of trust and safety.
Independently verified
SOC 2 Type II audit complete
FERPA‑aligned data handling
Age‑gated 13+, consistent with COPPA
Annual third‑party penetration test
Built‑in moderation
Moderates student input and flags what needs attention
Full visibility into conversation transcripts for teachers and admins
Real‑time self‑harm alerts
Protects data
No Playlab data is used to train third‑party AI models
US‑based, encrypted at rest and in transit
School sign‑in through Clever or Google
Building in Playlab? Name these safeguards in your team’s Safety section. It’s worth 10 points.
Part 2
AI’s limitations, risks and potential for upside
Guiding question: Can, and should, AI do it?
Can and should AI do it?
AI has great potential and real risks.
01Accuracy & fidelityDoes the output match what was asked and reflect factual reality?
02Bias & representationDoes it portray people and groups fairly, without stereotypes or default assumptions?
03AppropriatenessIs it free of harmful content and right for the students’ age?
04Educational valueDoes it support learning goals without creating confusion?
05Cultural sensitivityDoes it represent cultures, traditions and religions accurately and respectfully?
06Safety & misuseCould it cause harm if shared, taken out of context or misused?
07Technical qualityIs it clear, properly rendered and free of AI errors?
08AccessibilityCan every student use it, including students with visual or cognitive differences?
09Privacy & dataDoes it handle student information with care and collect only what it needs?
Can and should AI do it?
Why we need educators
Even in an AI‑native future
What AI can’t do
✕ Build student relationships
✕ Connect with students’ families
✕ Inspire intrinsic motivation
✕ Adapt to unexpected moments
✕ Give nuanced feedback on its own
What educators do
✓ Foster social‑emotional growth
✓ Create positive learning spaces
✓ Mentor and guide each student
✓ Build critical, creative minds
✓ Adapt lessons in real time
✓ Teach from content expertise
What’s changed
AI development timeline
1950s–80sRules
People write every condition by hand
1990s–2000sMachine learning
Give it examples, it finds the patterns
2010sDeep learning
Layered networks make speech and vision usable
2020sGenerative AI
Output becomes new text, not a label
NowAgents
Tools, memory and a goal, taking several steps on their own
What’s changed
2026: AI is now doing the research.
Click a card to learn more.
Before 2026: AlphaFold wins a Nobel Prize (2024) · Math Olympiad gold (2025) · AI‑designed antibiotics (2025)
May 2026 · OpenAI
80‑year‑old problem solved
What happened
An OpenAI model disproved Paul Erdős’s 1946 “unit distance” conjecture in geometry, given only the problem statement. Outside mathematicians reviewed the proof.
Why it matters for schools
AI can now produce new, checkable knowledge, not just summaries. What do students need to learn when the machine can do original math?
In October, OpenAI released 722 AI‑written manuscripts touching 100+ open problems. Many have computer‑checkable proofs, and OpenAI warns some may contain errors.
Why it matters for schools
Volume is outpacing human review. The same “who checks the work?” question shows up with AI‑generated lessons and feedback.
OpenAI published a claimed proof on the Navier–Stokes Millennium Prize problem, produced by thousands of coordinating agents. The Clay Institute is reviewing it, and credit is publicly disputed.
Why it matters for schools
A big AI claim isn’t the same as a verified result. A current example for teaching students to separate hype from evidence.
For schools: agents will soon act for students and staff. Who checks their work?
Jan–Mar 2026 · Moltbook
A social network for bots
What happened
Millions of AI agents post and reply to each other while humans watch. Meta bought it in March 2026. Earlier, a security flaw exposed private messages and over a million credentials.
Why it matters for schools
Agents are acting on their own in public spaces, and security often lags behind. Students will meet these systems before the rules exist.
A2A is an open standard that lets agents from different companies hand tasks to each other. The Linux Foundation reports 150+ supporting organizations, including Google, Microsoft and AWS.
Why it matters for schools
Soon one request may pass through several companies’ agents. Where does student data go, and who is responsible when something breaks?
During a security test, Google’s Gemini broke into three real companies’ systems after a setup mistake gave it internet access. Other labs’ models reportedly did similar things.
Why it matters for schools
Even the builders can’t fully predict agent behavior. Any agent acting for students or staff needs limits and human oversight.
Who do you think built it, and what do they care about?
You’ll need a news article
Open any article, then copy and paste its text into the app.
Ctrl/⌘ + A select allCtrl/⌘ + C copyCtrl/⌘ + V paste
Click Launch
Explore
Who built CivicLens?
Meet the builders!
Team Business Buds, three high school students in Fairfax County, VA, built CivicLens in Playlab.
“Right now in the US, an app like CivicLens feels like just what we need to help turn down the heat a bit and think more critically about the news we consume.”
Dr. Joseph South· Chief Innovation Officer, ISTE+ASCD
US Presidential AI Challenge winner
ISTE 2026 AI Challenge winner
Fairfax County Public Schools finalist
Explore
In edit mode · 5 min
Find it, then try it
Open CivicLens and click Edit. How many can you find?
Model selectionSwitch the model, then ask the same question. Did it change?
InputsAdd an input field users fill in, like grade level or topic.
MemoryMake the AI remember one specific thing, like your school’s name.
ConversationsFind where you can view past conversations with the app.
ExportFind how a result gets exported to Google Drive.
StandardsConnect the app to curriculum or state standards.
VoiceTurn on voice input or output, then talk to the app.
MathGet the app to take or show a properly formatted equation.
SharingFind every way to share your app with others.
Click Edit
Explore
Anatomy of an AI application.
What did you find under the hood? What goes in the middle?
User prompt
→
?
+
?
+
?
→
Output
Explore
Anatomy of an AI application.
User prompt
→
System promptThe recipe
+
Large language modelThe prediction engine
+
ReferencesYour expertise
→
Output
Part 2
Build: give direction to AI
Guiding question: How do we give direction to AI applications?
Build
Two prompts, one conversation.
System prompt
The standing instructions behind the app: role, rules and tone. Determines how the AI behaves.
Set before anyone arrives, invisible to the user.
User input
What the user types in the moment: the question, the request, the data, the follow‑up.
New with each message.
Build
The system prompt is your recipe. Write a detailed one!
01
Define the knowledge and expertise the AI application should have
02
Define the role of the AI application
03
Define the user and their context
04
Define success and the parameters of a strong output
05
Define the steps the app should follow, one by one
Build
Start with the problem
Build to solve a problem, for a specific user.
The strongest challenge projects start with a real person and a real problem in an Idaho school or community.
Example user
High school students in a rural Idaho town applying for their first job
Example problem
Few nearby adults to practice interviews with, and no way to get feedback before the real thing
Ask your team
Who exactly is the user?
What’s hard for them right now?
Why is AI a good fit for this problem?
Build
The system prompt is your recipe.
Background
This app has expertise in
This app’s role is to
The primary user of this app is
Success looks like
Workflow
1. First,
2. After they respond, then
3. Next,
Define what the app needs to know
Define the role of the AI application
Define the user and their context
Define success and a strong output
Define the steps, one by one
Your turn: draft one line for each blank.
Build · References
References add custom expertise
What the LLM already knows
What your app uses to produce the output
References
Your curricula and standards
Your rubrics
Research on your problem
Interviews with your users
Your community’s context
Build
Your turn
Build your first Playlab app.
1
Decompose a problem into a system prompt for an AI app.
2
Embrace play and progress over perfection. Build, test, iterate and experiment.
3
Need ideas? Ask the Idea Bot.
Build
Help while you build
Introducing the CoBuilder
The Playlab CoBuilder helps you build and improve your apps: troubleshooting, generating content or getting unstuck.
Try asking
“How can I make this workflow clearer?”
“What’s missing from this prompt?”
“The app isn’t doing X. How do I get it to do X, and where should it go?”
Build
Sharing your app
You choose how you want to share it.
Space
With your teammates
Playlab Community
For educators everywhere
Public link
That anyone can use
Build
Big idea
AI is designmaterial.
Part 2
Set up your space
Guiding question: How do we get our team building in Playlab?
Set up your space
Step 1
Create a space for your team
A space is the home for a class, team or group. It holds members, apps, activity and permissions.
1
In the sidebar under Projects, click Spaces
2
Click New space in the top right
3
Name it so students recognize it
4
Set the four member permissions
5
Click Create new Space, then Add apps
Don’t see New space? Ask your org admin.
Create new Space
Space name
Meridian HS · SI Challenge Team
Description
Our 2027 Standard Track project
Build Apps
Members can see each other’s apps
Share apps outside the space
Manage own apps
CancelCreate new Space
Set up your space
Step 2
Get your student invite link
1
Build your space and add the apps students will use
2
Turn on Build Apps, and leave Manage own apps off
3
Email rebekha@playlab.ai with your space name, or flag me during today’s session
4
I’ll check the setup and send your student invite link
5
Share the link with your students
Why this way
A quick safety check
Student invite links need top‑level org permissions. A quick review makes sure every space is set up safely before students join.
We can do this live today.
Set up your space
Step 3
What students do to join
1Open the inviteClick the link, or go to playlab.ai/join and enter the code
→
2Create an accountName, email and password, and confirm they’re at least 13
→
3Choose StudentPick Student as the role and add their school
→
4Land in the spaceThey see every app you’ve shared there
Heads up
With a link or code, students should fill in the invite form. Starting with Continue with Google fails until their account exists.
Part 2 · Next steps
Part 2 wrap‑up
So, what’s next?
Get your space ready this week so your team can start building.
1
Finish setting up your spaceBuild Apps on, Manage own apps off
2
Email ReBekha for your student invite linkrebekha@playlab.ai
3
Start students on the AI Impact Challenge courseIn class or on their own
4
Build a first version and test it with real usersThen try to break it on purpose
5
Bring what you learn into the proposalSolution, Safety and Learning Achievements
6
Bring questions to office hoursOct 27, 5–6pm MT
Closing
Questions? Let’s build.
Email rebekha@playlab.ai
Idaho playlab.ai/idaho-student-ai-challenge ↗
National si.gov/initiatives/presidential-challenge ↗
Appendix
What changed from last year
For teams that took part last year: how the 2027 rules differ.
Appendix · What changed
The name
AI
SI
“Artificial Intelligence” is now “Super Intelligence.”
A September 29, 2026 executive order, “Inaugurating the Era of Super Intelligence,” made the change across the executive branch. It’s now the Presidential SI Challenge.
New homes
Website
si.gov
White House
whitehouse.gov/edsi
Email
SI.Challenge@doe.gov
Appendix · What changed
Deadlines
About a month earlier and much tighter.
Support
Office hours on set dates Oct 19 – Dec 21, 7pm ET, for registered teams. Students attend with their Team Leader.
First Team Leader info session was Oct 14, 2pm ET
Milestone
2025–26
2027
Registration opens
Aug 26, 2025
Oct 13, 2026
Registration deadline
None separate
Dec 18, 2026, 5pm ET
Submission deadline
Jan 20, 2026, 11:59pm local
Dec 23, 2026, 5pm ET
State champions announced
Mar 16 – Apr 1
Feb 11–15, 2027
Regionals
Mar 27 – Apr 13
Mar 4 – Apr 8, 2027
Nationals (DC)
June 7–10, 2026
June 6–9, 2027
5pm ET is 3pm MT and 2pm PT for Panhandle teams.
Appendix · What changed
Eligibility
Everyone competes. Everyone shows proof.
Every student, educator and Team Leader must be a US citizen, US national or green‑card holder, and must show proof at each level.
2025–26
Last year
Teams could submit just for a participation certificate
Competition was a separate opt‑in
Only the competition required citizenship, national status or a green card
2027
This year
Every eligible submission is scored by at least three judges for the state round
Citizenship, national status or a green card is required for everyone
Proof is checked at state, regional and national levels
Appendix · What changed
Tracks
The tracks were restructured.
Track I · Proposalall students
→
Standard TrackAll grade levels, solving a problem for students, families or communities
Track II · Technical buildall students
→
Trailblazer Track NewGrades 6–12 only. New or unique projects addressing a national challenge, with an eligibility check on originality and national reach
Track III(a) Teaching AI Track III(b) AI toolseducators
→
Educator CategoryOne combined category
Elementarywhole classes or groups
→
Elementary1–5 named students
Appendix · What changed
Written submission
From one essay to a structured proposal.
2025–26
Last year
One PDF: up to 50 MB, 10 pages max, 12‑pt font+
Team narrative of at least 500 words, citing all tools and sources
No required headings, just questions to “consider”
Signed Certification of Originality
2027
This year
One PDF: up to 30 MB, 10 pages plus a title page, 12‑pt font+
No word minimum
Required sections for every submission
Everything original and created after Aug 1, 2026
Appendix · What changed
Framing
It’s a proposal now.
Teams don’t need a fully working build to submit. The video shows how feasible and promising the idea is. The finished tool isn’t due until nationals.
Framed as a proposal
Last year Track II teams wrote about something they had already built. Now every student team writes a proposal.
Originality is a timing rule
No separate Certification of Originality. Everything must be original and created after Aug 1, 2026.
Educator questions combined
III(a) and III(b) are one list, “not all may apply.” Educators are also asked about scalability and whether the idea could become a best practice.
Appendix · What changed
Media
Every track needs a video.
2025–26
2027
Standard(was Track I)
Poster link (digital or photo)
Video
Trailblazer(was Track II)
Video or demo, 4 min max
Video or demonstration
Educators
Video or demo, 4 min max
Video or demo
4:00
Judges stop watching at 4 minutes.
Under 300 MB
Linked from the title page
Scored by judges: 10 pts
No password on the link
Test the link on several different computers
Credits count toward the 4 minutes
Appendix · What changed
Scoring
All rubrics are out of 100.
Removed
Relevance to the President and his Administration(10 pts, first tiebreaker)
New
Safety10 pts in every rubric
New
Video10 pts in every track
But the priority moved into the framing
Projects should address “challenges of concern to President Trump, the First Lady, his Administration, and communities across the Nation.”
About the Challenge, 2027 guidebook
New tiebreakers
Standard
Originality, then SI Solution
Trailblazer
Research Supporting Novelty, then SI Solution
Educator
Innovation, then Objective
Appendix · What changed
Prizes
Smaller, and spread across more levels.
Level
2025–26
2027
Participation
Presidential Certificate
Electronic certificate by Aug 31, 2027
State
Certificate + cloud credits and other partner perks
$200 gift card per person
Regional
Certificate + perks + possible DC invitation
Invitation to DC finals + $1,000 gift cards ($1,500 Trailblazer)
National
$10,000 per member (elementary: $10,000 to the school)
Elementary $2,500 · Middle $7,500 Standard / $10,000 Trailblazer · High $10,000 / $12,500 · Educators $10,000
Appendix · What changed
Eligibility
Past winners and regions
2026 regional champions
Middle and high schoolers can’t compete in the same category again
Elementary students can join only school‑based teams
Educators can lead student teams but can’t join an educator team
All are invited to a new Ambassadors Program
Map changes
DOWEA and Tribal schools each get their own state‑level champion
Two‑champion states drop from 7 to 4: CA, FL, NY, TX