2024 Xhope TAI Institution Design Hackathon

2024 Xhope TAI Institution Design Hackathon

To help AI development benefit humanity, Foresight Institute has held various workshops over the past years, such as AGI & Great Powers, AGI: Toward Cooperation, in addition to two technical meetings in 2022 and 2023 focusing on Cryptography, Security, and AI. Recently, we also launched a Grants Program that funds work on AI security risks, cryptography tools for safe AI, and beneficial multipolar AI scenarios.

While these and other efforts have shed light on a diversity of promising projects to support, the intersection between the AI domain and the cryptography and security domain remains nascent but promising. This is why our upcoming workshop “AGI: Cryptography, Security & Multipolar Scenarios” invites leading researchers, entrepreneurs, and funders to collaborate across fields and explore tools and architectures that help humans and AIs cooperate.

Questions you will explore

1. Which challenges in AI alignment, AI security and AI coordination (in particular for multipolar AI scenarios) are critical to address but have not yet been sufficiently addressed? 
2. Which tools in cryptography, security, game theory, mechanism design, and auxiliary fields are uniquely positioned to make progress on these challenges? How can we bridge-build across fields? 
3. In addition to addressing these challenges, are there potential novel beneficial applications that could be unlocked by those tools? Which prototypes can we build now?

Themes you will explore

• Computer security to help with AI infosecurity or approaches for scaling up security techniques to potentially apply to more advanced AI systems
• Cryptographic and auxiliary techniques for building coordination/governance architectures across different AGI(-building) entities
• Privacy-enhancing technologies for improved verification/evaluation techniques
• Game theory and simulations for multi-agent scenarios
• Mechanisms for avoiding collusion and deception, and incetivizing pareto-preferred and positive-sum dynamics
• Approaches for tackling principal agent problems in multipolar systems

Format

Participants are invited to explore new opportunities, form lasting collaborations, and further cooperation toward shared long-term goals. In addition to short presentations, working groups, and project development, we offer mentorship hours, open breakouts, and speaker and sponsor gatherings.