AISI Policy Hackathon

Are you ready to shape the future of AI governance alongside industry leaders and renowned experts? The inaugural AISI AI Policy Hackathon is your gateway to making a real impact on urgent challenges in AI governance while building career-defining connections. This hackathon is open to GT students from all majors, both undergrad and graduates.   Join us from April 5-6 to compete for up to $1,000 in prizes and automatic acceptances into our fellowship!

Logistics

Day 1: April 5 (In-person at Klaus 2443, food provided!)

• 11:00 AM: Welcome & Check-in

• 11:30 AM - 12:15 PM: Workshop 1

• Prof. Charlotte Alexander on "Policy Brief Development & Stakeholder Analysis"

• Join the zoom here, open to ALL (not just participants!)

• 12:15 PM-12:45 PM Lunch

• 12:45 PM - 1:30 PM: Workshop 2

• Michael Chen, Member of Policy Staff @ METR (featured in The Guardian, MIT Tech Review, and Financial Times)

• Join the zoom here, open to ALL (not just participants!)

• 2:00 PM: Workshop #3 (Full details coming soon!)

• 3:00 PM onwards: Networking session + Remote office hours

Day 2 (April 6) - Virtual Sessions

• Virtual mentor office hours + work with your team to develop solutions across three critical tracks:

  • International Cooperation

  • Deployment & Application Regulation

  • Training Data Privacy & Copyright

Final Presentations: April 13, 2025

Prizes

• 1st place team: $500 + acceptance to AISI fellowship

• 2nd place team: $300 + acceptance to AISI fellowship

• 3rd place team: $200

Deliverables

• Teams will submit policy briefs addressing challenges in their chosen track:

  • Maximum 10 pages (excluding appendices and references)

• Addressed to a relevant regulatory agency or governing body (e.g., National AI Initiative Office, FTC, EU Commission, etc.)

• Recommended components

  • Executive Summary (1 page)

  • Problem Statement & Background (1-2 pages)

  • Recommendations & Implementation Plan (4-5 pages)

  • Impact Assessment (1-2 pages)

  • References & Appendices

Evaluation criteria

• Relevance and Impact (20%)

• Innovation and Creativity (20%)

• Argument sophistication (20%)

• Feasibility (20%)

• Presentation and Documentation (20%) 

✨ Why AI Governance Matters Now

The rapid advancement of AI technologies presents unprecedented challenges for governance frameworks worldwide. As systems become more capable and widespread, the need for thoughtful policy approaches has never been more urgent. This hackathon addresses critical gaps in our current regulatory landscape by empowering participants to develop innovative, actionable solutions.  We encourage interdisciplinary teams! While helpful, prior policy experience is not required—we'll provide resources and mentorship to guide participants at all experience levels.

🎯 Challenge Tracks

See this document for full details, ideas, and resources!

1. International Cooperation

Bridge the divides in AI regulation across national boundaries. Work on frameworks that facilitate global collaboration while respecting regional differences in values, priorities, and regulatory approaches.

Key questions to explore:

• How can we harmonize AI governance standards across jurisdictions?

• What mechanisms can foster agreement between different political systems?

• How might we design adaptive international frameworks that evolve with AI capabilities?

2. Deployment & Application Regulation

Develop policies ensuring safe and ethical AI operation in high-stakes domains like healthcare, transportation, and public safety. Focus on proportional oversight mechanisms that scale with risk levels.

Key questions to explore:

• What testing and certification processes should precede AI deployment in critical sectors?

• How can we balance innovation with appropriate safety precautions?

• What ongoing monitoring frameworks should be implemented for deployed systems?

3. Training Data Privacy & Copyright

Design balanced approaches to data rights that protect individual privacy and intellectual property while enabling beneficial innovation.

Key questions to explore:

• How should we reform copyright law to address AI training data concerns?

• What compensation models might work for creators whose content trains AI systems?

• How can we ensure transparent data practices while protecting proprietary interests?