Third-party cyber evaluations involving OpenAI models
openai.com·13h ago
AI Research
arXiv CS.AI·May 26·5 min read
Is Decentralized AI Governable? From Regulative Policy to Constitutive Protocol
TL;DR
Decentralized AI (DeAI) creates a governance challenge because it lacks identifiable entities responsible for compliance. The authors propose a shift from traditional regulatory policies to protocol-based governance, which focuses on architectural constraints rather than normative rules.
✦ Why It Matters
Engineers and researchers can adopt protocol-based governance to enhance accountability in decentralized AI systems.
Key Takeaways
- Understand the six-layer stack of decentralized AI governance.
- Recognize the accountability and incapacitation gaps in DeAI.
- Implement protocol-based governance to shape AI actions.
- Ensure ethical conditions are met to avoid technocratic power.
How It Works
The authors propose that governance should shift from normative policies, which rely on identifiable agents, to protocols that define the architecture of AI systems. This approach focuses on shaping the environment in which AI operates, thereby influencing the types of actions that can be taken without needing to identify a responsible party.