2026 AI Safety Science Report
Early signals from the 2026 AI Safety Science Report on model-level cost asymmetry, induced behaviours, scalable oversight and system-level risk propagation.
-
Read more: 2026 AI Safety Science Report -
Unlocking the Future: Scaling Responsible/Safe AI for 2026
Read more: Unlocking the Future: Scaling Responsible/Safe AI for 2026CSIRO Data61’s 2026 research directions in scalable oversight, meaningful human agency and cross-domain techniques that jointly advance AI safety and economic value.
-
Rethinking Risk for AI Generated Content
Read more: Rethinking Risk for AI Generated ContentWhy AI-generated content guidance should assess potential impact and degree of AI involvement, rather than relying on unstable likelihood estimates or nominal human oversight.
-
What public auditing can teach us about governing AI
Read more: What public auditing can teach us about governing AIA CSIRO Data61 partnership with the Audit Office of NSW explores responsible AI for public auditing and what established audit practice can teach AI governance.
-
International AI Safety Report: Second Key Update 2025
Read more: International AI Safety Report: Second Key Update 2025As part of the international expert advisory group representing Australia, I’m pleased to see the second Key Update of the AI Safety Science Report, led by Yoshua Bengio, released (link in the comment). The official update captures the broad progress: more robust adversarial training, better AI-generated content tracking, wider adoption of frontier safety frameworks, and…
-
Australia’s Next Step in Responsible AI Adoption
Read more: Australia’s Next Step in Responsible AI AdoptionEvery few weeks brings a new AI framework — but rarely one that genuinely helps organisations simplify. The National AI Centre‘s new guidance on responsible AI adoption, launched this week, is a welcome exception. It distils earlier guidance and complex international work into six clear, balanced practices that bridge governance intent with operational reality. We…
-
AI & Health, Safety, and Environment (HSE)
Read more: AI & Health, Safety, and Environment (HSE)Cognitive overload, opaque automation, and AI surveillance are reshaping the modern workplace. I’m sharing some insights from my talk at the Safety Sphere, Australia’s leading HSE practitioner network, on “𝗪𝗵𝗲𝗻 𝗔𝗜 𝗕𝗲𝗰𝗼𝗺𝗲𝘀 𝗮 𝗖𝗼𝘄𝗼𝗿𝗸𝗲𝗿: 𝗨𝗻𝗱𝗲𝗿𝘀𝘁𝗮𝗻𝗱𝗶𝗻𝗴 𝗛𝗦𝗘 𝗥𝗶𝘀𝗸𝘀 𝗮𝗻𝗱 𝗢𝗽𝗽𝗼𝗿𝘁𝘂𝗻𝗶𝘁𝗶𝗲𝘀 𝗼𝗳 𝗚𝗲𝗻𝗲𝗿𝗮𝗹-𝗣𝘂𝗿𝗽𝗼𝘀𝗲 𝗔𝗜.” We examined how AI not only enhances productivity but also creates new health,…
-
Representing Australia in the Global Effort for AI Safety
Read more: Representing Australia in the Global Effort for AI SafetyIt has been a great honour to represent Australia on the Expert Advisory Panel for the International AI Safety Report, led by Joshua Bengio and backed by over 30 countries and international organisations including the EU, OECD and UN. This is the most important ongoing effort to monitor AI capability gains and safety risks of…
-
Future-Proof Protection for Sensitive Data
Read more: Future-Proof Protection for Sensitive DataCSIRO’s Data61 latest research gives copyright owners and data producers a new superpower — the option to make their sensitive and sovereign data 𝘂𝗻𝗹𝗲𝗮𝗿𝗻𝗮𝗯𝗹𝗲 by AI. It’s been getting some attention, with us appearing on ABC TV, 7 News, and 2GB radio to explain the technology. Unlike past “confuse the AI” tricks like data poisoning,…
-
IEEE Software Best Paper Award
Read more: IEEE Software Best Paper Award🏆Our paper — “Toward Responsible AI in the Era of Generative AI: A Reference Architecture for Designing Foundation Model-Based Systems” has been awarded the Best Paper Award of IEEE Software 2024! https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=10553223 Although CSIRO’s Data61 often publish breakthrough research and win best paper awards at academic conferences, nothing beats being recognised by practitioners. The world-leading…
