📋Use Case Information
Impact Scale: Consider potential consequences if the AI system fails or produces incorrect results.
Likelihood Scale: Assess probability of system issues based on complexity and validation.
🌱Carbon Footprint Assessment
Model Size: Larger models require more computational resources and energy.
Training Frequency: More frequent training increases energy consumption.
Infrastructure: Green energy sources significantly reduce carbon footprint.
Usage Volume: Higher inference volumes increase overall energy consumption.
🛡️Governance & Security
Privacy Risk: Higher privacy risks require enhanced governance controls.
Security Level: Poor security increases operational risk significantly.
AI Safety: Poor alignment increases risk of unintended consequences.
Regulatory: Non-compliance can result in severe penalties and deployment blocks.