What does the international scientific community see as the biggest opportunities, risks and governance priorities for AI?

The newly released Preliminary Report from the UN Independent International Scientific Panel on AI provides one of the most comprehensive evidence-based answers to that question to date: https://lnkd.in/gYSZxUpy

Congratulations to my colleague Qinghua Lu for leading the working group on Security, Systems and Environmental Implications. These are among the most important questions facing the development and deployment of AI systems.

I was also pleased to see CSIRO‘s research recognised in the report, including citations to our work:
• Zhu, L., Lu, Q., Ding, M. et al. (2026). Designing Meaningful Human Oversight in AI. AI Ethics, 6, 286. https://lnkd.in/gUyCSeCf
• Zhu, L., & Lu, Q. (2026). Verifiability-First AI Engineering in the Era of AIware. SSRN. https://lnkd.in/gQsaNyUV

Our works do not stop at identifying risks or proposing better risk evaluation methods. They focus on concrete AI engineering solutions for building trustworthy AI systems. We are now working closely with Australian industry to operationalise these research breakthrough and technologies.