Verification Mechanisms
Tools and protocols for proving whether AI systems meet safety claims across training, evaluation, and deployment.
Research
We make Singapore a proving ground for technical AI governance and a place where research can move across regional and institutional lines.
Our Approach
We work to make Singapore both a proving ground for technical AI governance and a bridge between the global communities building it. Our research runs from agent governance to cyber misuse to frontier evaluations.
Reports, demos, and project sites are built for use by researchers, labs, and public institutions.
Featured Research
Recent publications and working papers from SASH and our collaborators.
Detecting Offensive Cyber Agents
AI agents can now orchestrate cyberattacks, changing the speed and scale of cyber threats. This report proposes detection-in-depth as a strategic framework for defenders and policymakers.
Agent IDs
Agent IDs help services and people understand who controls an AI agent, what it is permitted to do, and what should happen if something goes wrong.
Frontier AI Safety in Asia
Neo Research is an independent research and evaluations lab built in and for Asia, focused on ensuring that frontier AI can be safely deployed at scale.
Research Focus
Frontier Safety
Risk analysis and safeguards for the most capable models before they move into high-stakes public and commercial use.
Agent Governance
Frameworks for identifying, supervising, and constraining autonomous agents as they act on behalf of people and institutions.
Verification Mechanisms
Methods for checking and comparing safety claims across training, evaluation, and deployment.
Frontier Safety
Research on frontier model risks and safeguards before high-stakes deployment.
Agent Governance
Governance patterns for identifying, supervising, and constraining autonomous agents.
Interested in collaborating on research with us?
Reach out to SASH about research collaboration.
Email SASH