DSCI Floats Framework To Determine Liability For Harm Caused By AI Agents
The DSCI has released a discussion paper proposing an industry framework to determine liability when autonomous AI agents cause harm. Rather than creating new regulations, it recommends applying existing laws and sectoral rules to agentic AI systems. The framework assigns responsibility based on control and decision-making across five participant groups: end-user organisations, service providers, product companies, tool providers, and infrastructure providers. The shift to agentic AI—where systems independently plan tasks, access APIs and execute instructions with minimal human oversight—poses new accountability challenges that the framework seeks to address.
Key points
- DSCI proposes liability framework for AI agent harm
- Responsibility assigned based on control across value chain
- Applies existing laws rather than new regulations
- Addresses risks from autonomous agentic AI systems
Summary and key points written by AI from the headline and a short excerpt, so they can be incomplete or wrong. Inc42 did the reporting; read the original for the full story.
Read the full story at Inc42
