The International Network for Advanced AI Measurement, Evaluation and Science is the new name for the International Network of AI Safety Institutes. We met for the fourth time on 4 and 5 December in San Diego.
The network’s name change reflects a focus to work together on a rigorous, shared approach to AI measurement.
AI evaluation science is becoming a critical, collaborative area of focus that brings together industry, civil society, academia and governments. All these areas are working to develop reliable methods for assessing AI systems and foster shared standards.
To support this focus on AI measurement and evaluation, the network held several workshops to discuss best principles and approaches to conducting AI evaluations.
Australia’s participation in the network convening took place a week after the Australian Government announced it would establish an Australian Artificial Intelligence Safety Institute (AISI). The AISI will start to operate in early 2026 to support the government’s ongoing response to emerging risks and harms associated with AI technologies.
Australia joined network members from:
- Canada
- the European Commission
- France
- Japan
- the Republic of Korea
- Singapore
- the United Kingdom
- the United States.
The UK AI Security Institute will take on the role as Network Coordinator for the next 12 months.
The work of the network builds on the Seoul Statement of Intent toward international cooperation on AI safety science. Released following the AI Seoul Summit on 21 May 2024, it confirms a shared understanding of the opportunities and risks posed by AI.