Grants for AI in Education 2026: Win Part of $400,000 from the Stanford Create+AI Challenge

If you build tools, teach students, research learning science, or design edtech experiences, this one is worth your attention. Stanford’s Accelerator for Learning and Google.org are running the Create+AI Challenge 2026, a prize program that will distribute $400,000 across awards for projects that use artificial intelligence to augment teaching, learning, and career pathways—with people, not automation, at the center.

This is not a contest for flashy demos alone. The judges want projects that show respect for how humans teach and learn: thoughtful measurement, accessibility, and real classroom or community impact. Expect money, mentorship, and stage time at an influential summit—plus the chance to develop further at Stanford in summer 2026. If you can assemble a tight pitch, a two-minute video, and a thoughtful plan grounded in learning science, you could walk away with $10K–$50K and a support network that matters.

Below I break down who should apply, how to prepare an application that reviewers will remember, and a realistic timeline to get everything polished before the January 12, 2026 deadline.

At a Glance

Why This Opportunity Matters

Artificial intelligence in education often gets boxed into automation or test scoring. The Create+AI Challenge bets on a different approach: using AI to amplify what educators and learners already do well. That might mean tools that free teachers from paperwork so they can build relationships, assistive technologies that let students with learning differences participate fully, or platforms that connect learning to real career paths.

Beyond the money, the value is in access to expertise. Winners receive mentorship from Stanford faculty and researchers, invitations to co-design workshops, and visibility at a high-profile summit. For early-stage teams, those connections can shorten research cycles and open doors to pilots, partners, or follow-on funding.

Lastly, the track structure helps you target where your work fits. Whether your project supports teachers, learners, or career transitions, the call is explicit: show how AI supports human talent and interaction. If your project treats AI as a helpful colleague rather than a replacement, you’re in tune with the program’s spirit.

What This Opportunity Offers

This challenge distributes $400,000 across several awards. The headline prizes are two $50,000 awards in each track—Augment Teaching, Augment Learning, and Augment Career Opportunities. On top of those, multiple smaller awards ($10K–$20K) will recognize promising projects across the tracks.

Funding is only part of the package. Winners get curated mentorship from Stanford researchers, access to design workshops that center learning science and teacher co-design, and exposure at the AI+Education Summit on February 10–11, 2026. That summit is an ideal stage: you’ll meet potential partners, pilot sites, and funders who care about evidence-based design.

The organizers also may invite teams to join a summer 2026 development cohort at Stanford. That’s not guaranteed funding, but it’s an opportunity to accelerate product or research development with institutional support. In short: cash, credibility, coaching, and an audience. If you’re serious about scaling responsibly and measuring learning impact, this combination is rare and useful.

Who Should Apply

This challenge is intentionally broad: educators, researchers, designers, nonprofits, entrepreneurs, and students are encouraged to submit. But the program has clear expectations about orientation and capacity.

If you’re a teacher or school leader with a small team and a prototype that reduces admin burden or improves classroom relationships, apply. If you’re a researcher with a lab tool that measurably improves engagement for learners with disabilities, apply. If you run a nonprofit building mentorship and career pipelines and want to incorporate AI to personalize guidance at scale, apply.

A practical constraint: at least one team member must have Stanford affiliation (current student, scholar, staff, or alum). That’s a barrier for some international teams, but it’s also a bridge: you can partner with a Stanford alum or researcher. Another option is to collaborate with Stanford students through incubators or faculty you know—many faculty are receptive to co-design partnerships that give their students real-world projects.

Examples:

• A Kenyan NGO with an edtech prototype partners with a Stanford alum who can represent the team’s Stanford connection and shepherd ethics and research design.
• A university researcher with pilot data on an adaptive reading tutor teams up with a teacher-coach and a UX designer to create an ambitious, classroom-ready submission.
• A startup that provides career coaching prototypes an AI mentor to scale mentorship; they include an ex-Stanford employee on the team to meet the affiliation requirement.

If you are a Stanford employee, confirm participation complies with university policies regarding outside activities, conflict of interest, and use of university resources before applying.

Selection Criteria — What Judges Care About

A review panel of educators, researchers, technologists, and learners will score proposals on several criteria:

• Innovation and creativity: Is the idea original within its context?
• Learning impact: Will it measurably improve learning outcomes or participation?
• Fairness and inclusion: Does it address equity, accessibility, and bias?
• Use of learning sciences and design: Is the approach grounded in evidence or sound learning theory?
• Measurement plan: Are outcomes clearly defined and measurable?
• Feasibility and sustainability: Can this be carried out and maintained beyond the award?

Think of these as a checklist you should answer clearly in your proposal. Judges want novelty, yes, but novelty without evidence or a measurement plan won’t score highly. Likewise, a small but rigorously measured pilot can beat a flashy demo with no evaluation plan.

Insider Tips for a Winning Application

1. Tell a human story first. Begin your pitch deck and the two-minute video with a clear, concrete user story: a teacher, student, or job seeker who benefits. That single narrative helps reviewers quickly grasp impact.

2. Make fairness explicit. Address accessibility, bias mitigation, and how the AI handles underrepresented groups. Don’t rely on vague statements—show steps you’ll take (data audits, diverse user testing, alternative input modes).

3. Define outcomes that matter and are measurable. Don’t promise “better learning.” Promise a specific, observable metric—improved mastery on a formative assessment, increased participation rate among students with disabilities, or a measurable change in career-readiness indicators—plus the tools you’ll use to measure it.

4. Keep the proposal tight and visual. The PDF is limited to 10 pages. Use one or two well-designed figures: a logic model showing inputs→activities→outcomes, and a timeline with milestones and measurable checkpoints. Visuals clarify faster than long paragraphs.

5. Use the two-minute video to add personality, not technical detail. Show the team, show the prototype in action (even a quick screen recording), and end with a sharp statement of the problem and your solution’s benefits. Make it human and confident.

6. Be realistic about scope and budget. If you ask for $50,000, show how that money directly enables the next critical step—pilot expansion, rigorous evaluation, or building a teacher-facing workflow. Break down costs in a short budget summary.

7. Plan for implementation from day one. Judges prefer projects that show awareness of real-world constraints: teacher time, connectivity limits, data privacy laws, and budgets. Include mitigation strategies and realistic partnerships for pilots.

8. Partner strategically to meet the Stanford affiliation requirement. If you don’t have a Stanford person on your team, recruit an alum or a student with relevant skills who can also help with research design. Offer clear roles and a short memo of understanding.

These tips together form an application that reads like a plan rather than a wish list.

Application Timeline (Realistic and Backward-Looking)

Deadline: January 12, 2026. Don’t leave this to the last week.

• December (4–6 weeks before deadline): Finalize core narrative, secure Stanford affiliate, draft video script, and start building visuals. Work with your institution or legal advisor if you need sponsorship letters or data agreements. Prepare a one-page budget.

• Mid December (3 weeks before): Record and edit the two-minute video. Complete the PDF proposal draft and circulate to at least three reviewers—one in education, one technologist, and one non-specialist. Incorporate feedback.

• Early January (1–2 weeks before): Final proofread, ensure the Google Form is complete, and upload PDF and video file links. Submit at least 72 hours early to avoid last-minute technical issues.

• January 12, 2026: Official deadline. Applications submitted after this are unlikely to be considered.

• February 10–11, 2026: AI+Education Summit visibility for selected teams.

• Summer 2026: Potential invited cohort for project development at Stanford.

Give yourself buffer days. Video uploads, collaborator sign-offs, and file formatting often cause delays.

Required Materials and How to Prepare Them

Your submission should include a completed Google Form and these attachments:

• PDF proposal or pitch deck (max 10 pages). Include project title and selected track, summary, problem and target audience, proposed solution and how it uses AI, connection to learning science, expected outcomes and measurement plan, fairness/accessibility considerations, timeline and milestones, team bios, and a budget summary. Write clearly and use visuals for complex ideas.

• Two-minute video introducing the idea. Keep it under 2 minutes. Use subtitles for accessibility. Show the human problem, the product or prototype, and the core team.

• (Optional) Prototype/demo or short video. A short screen recording or live demo can make a huge difference—especially if it shows the product working in context.

• (Optional) Financial model or budget. For teams requesting larger awards, include a simple budget spreadsheet showing personnel, development, pilot costs, and overhead.

Preparation advice: script your video before recording, and rehearse one or two takes rather than cobbling together clips. For the PDF, prioritize clarity—judges will scan for the problem, solution, evidence, and measurement plan. Use appendices sparingly and only if they add real support (e.g., preliminary pilot data).

What Makes an Application Stand Out

Reviewers reward applications that combine empathy, evidence, and execution. Standout features include:

• A clear problem-person-solution narrative that starts with a real classroom or workplace example.

• A simple but rigorous measurement plan tied to learning science concepts—pre/post assessments, validated instruments, or engagement metrics with defined thresholds.

• Demonstrated involvement of the people affected: teacher co-design sessions, advisory groups with learners, or pilot sites that can validate feasibility.

• Concrete fairness work: a plan for bias testing, translated interfaces, accessibility features, or inclusive datasets.

• A realistic plan for sustainability: partnerships with districts, nonprofits, or employers; cost models that show how the intervention can be scaled; and roles for local stakeholders.

If you can show small-scale efficacy plus a credible route to scale, you will be taken seriously.

Common Mistakes to Avoid (and How to Fix Them)

1. Vague Outcomes: Saying “improve learning” without specifics. Fix: define measurable outcomes and how you will measure them within your timeline.

2. No Stanford Connection Plan: Waiting until the last week to secure a Stanford affiliate. Fix: reach out early—contact alumni groups, relevant faculty, or student organizations to find a partner.

3. Overambitious Scope: Proposing a nationwide randomized trial in 6 months. Fix: scale your goals to the award’s size—propose a focused pilot with clear milestones.

4. Ignoring Accessibility or Bias: Treating fairness as an afterthought. Fix: include concrete steps—user testing with diverse groups, model audits, and alternative interfaces.

5. Weak Video: A two-minute video that repeats the PDF verbatim or feels scripted. Fix: show users, show the prototype in action, and make it human.

6. Poor Budget Justification: A line-item budget with no narrative. Fix: explain how each major cost contributes to outcomes—personnel for data collection, costs for pilot sites, or funds for accessibility work.

Address these common mistakes early; reviewers notice preparation and honesty.

Frequently Asked Questions

Q: Do I need to be affiliated with Stanford to apply? A: At least one team member must have a Stanford affiliation (current student, scholar, staff, or alum). Teams can include international members, but the Stanford affiliate is required. If you don’t have one, recruit a collaborator or contact alumni networks.

Q: Can teams include international members or be based outside the U.S.? A: Yes. International team members are allowed, subject to U.S. sanctions and export-control restrictions. Ensure legal and logistical considerations are clear in your plan.

Q: How long should the proposal PDF be? A: Maximum 10 pages. Use the space wisely—focus on problem, solution, measurement, fairness, timeline, team, and budget summary.

Q: What should the two-minute video show? A: Introduce the team, present the problem, show the prototype or mockup, and state expected outcomes. Keep it human, clear, and under two minutes. Add captions for accessibility.

Q: If we don’t get a top award, are there other benefits? A: Yes. Many smaller awards ($10K–$20K) will be granted. Plus, participating can get you noticed by mentors and summit attendees, which could lead to pilots or partnerships.

Q: Will winners get ongoing support? A: Winners receive mentorship and may be invited to a summer cohort at Stanford. There’s also visibility at the February summit for networking.

Q: How are proposals judged? A: Panels of educators, researchers, technologists, and learners evaluate innovation, learning impact, fairness, use of learning science, measurement, feasibility, and sustainability.

How to Apply (Next Steps)

Ready to start? Here’s a concrete checklist to move from idea to submission:

• Confirm you have a Stanford-affiliated team member. If not, recruit one now.
• Draft a one-page problem statement and a one-paragraph measurement plan.
• Script the two-minute video and plan a short demo or screen recording.
• Assemble a 6–10 page PDF pitch deck covering problem, solution, learning science, outcomes, fairness, timeline, team, and budget summary.
• Record and caption your video. Export as MP4 and host it on a sharable link if the form requires a URL.
• Submit through the Google Form well before January 12, 2026.

Ready to apply? Visit the official application page and complete the online form: https://docs.google.com/forms/d/e/1FAIpQLSfLsczrSwGPyj8PdL6z6UR1yOp-Rx3D9sTo_mHARgmvaCqSJw/viewform

If you want feedback on a draft pitch or video script, reach out to colleagues now and schedule at least one outside review from someone who is not deep in your subfield. Fresh eyes catch the things reviewers will notice.

Good luck—this is a fast-moving, high-value opportunity for teams that combine empathy, measurement, and practical design. If your AI idea helps people teach better, learn more, or enter meaningful work, this grant could help you get there.