Liam Carroll

Welcome and Opening Logistics

The 2026 Forum opens on the frame that shapes the next two days: the International AI Safety Report, chaired by Yoshua Bengio and released in February 2026. Frontier AI capabilities are rising quickly and diffusing through the economy just as quickly, bringing risks that run from cybersecurity and labour market disruption through to loss of control — none of which engineering alone will solve. Australia has the expertise to respond, but it sits scattered across disciplines and institutions that rarely meet in one room. This opening session sets out what the Forum is for: turning that dispersed expertise into a community that can act together. It also marks how fast the field is growing here, from around 100 people in the room in 2024 to close to 300 this year.

In conversation: Anna Goldsworthy on "The God We Made: The Threat and Promise of AI"

How will we be changed by the rise of artificial intelligence? In her recent Quarterly Essay "The God We Made", Anna Goldsworthy argues that AI is a rupture forcing us to confront what it is to be human, raising new existential, social and ethical risks while asking what remains irreplaceable in us. Anna joins Liam Carroll, an AI safety researcher at the Gradient Institute, in conversation about these questions.

Welcome and introduction to Day 1

Introduction to Day 2

Emerging practice in technical AI safety

This session explores concrete initiatives in technical AI safety through AI evaluations development, talent cultivation, and research acceleration. Join Soroush Pour from Harmony Intelligence and Ryan Kidd from MATS Research as they share their experiences building concrete AI safety projects and programs, followed by open discussion.

Concluding remarks