AI Alignment & Governance
Engineering AI alignment for a safer tomorrow.
LLM Alignment Benchmarking Dashboard
Risk Management
Transparency
Fairness
Safety
Robustness
Accountability
Detailed Metrics
| LLM | Risk Management (%) | Transparency (%) | Fairness (%) | Safety (%) | Robustness (%) | Accountability (%) |
|---|
The Artificial Superintelligence Cluster (ASIC)
The ASIC, is an MIT open-architecture that pioneers alignment engineering using open data and governance throughout the AI stack.
Project Overview
ASIC uses global standards found in finance, energy, academia and by governments around the world for alignment benchmarking via ISO/IEC 42001, NIST AI Risk Management Framework, and the EU AI Act.
AI Alignment
Objective
Our objective is to advance AI alignment by promoting top down transparency, ensuring pro-human AI collaboration for the next generation.
Approach
ASIC takes a three way approach to the alignment question by bridging engineering with data science and a public core governance layer.
Team
We are experts in government, academia, finance, and energy committed to driving transparent AI alignment.