(selected slides only) Responsible/Trustworthy AI in the Era of Foundation Models The emergence of large language models (LLM) such as GPT4 has garnered significant attention, placing foundation models at the forefront of AI systems. However, integrating foundation models raises concerns regarding responsible/trustworthy AI due to their opaque nature and rapidly moving capability boundaries. This talk…
Title: Software Engineering as the Linchpin of Responsible AI Abstract: From humanity’s existential risks to safety risks in critical systems to ethical risks, responsible AI, as the saviour, has become a massive research challenge with significant real-world consequences. However, achieving responsible AI remains elusive despite the plethora of high-level ethical principles, risk frameworks and progress…
With generative artificial intelligence (AI) and the pace of change dominating headlines across the globe, there is a growing focus on the importance of safe and ethical AI. Governments, businesses, and citizens all want to know how to ensure responsible AI is developed and deployed in a way that benefits business and community. Join the…
A significant shift seems to be occurring within the AI startup scene. How do you define your unique advantages when bespoke AI models can no longer effectively compete with Large Language Models (LLMs) and Foundation models? Could a larger team of product managers and a deeper understanding of your market and customers really provide a…
It was a great pleasure to join the panel online (with experts from US NSF, UK and EU) on International Cooperation for Research at the Privacy Symposium https://lnkd.in/gafct3DX. I presented CSIRO’s Data61 Australia’s latest research and regulation development in privacy and data protection and used responsible AI as an example of how we are forming a global research…
It was great to host the US NSF delegation (including the honorable Sethuraman Panchanathan) and minister Ed Husic MP yesterday and showcase the latest in CSIRO’s Data61 responsible AI research, focusing on responsible AI engineering that bridges the gap between high-level principles/frameworks/standards and low-level algorithms. In addition to showing how our responsible AI pattern catalogue https://lnkd.in/gkxVSKTz has been used in AU companies…
It was my pleasure to deliver the AICS 2022 keynote yesterday. You can find a subset of the slides below. As CSIRO’s Data61 engage with our industry and government partners, we often hear tales of disjointed and growing risk types (cybersecurity, AI, privacy, ethics, ESG, sovereignty.. ) and the lament of scarce expertise in each of them.…
I always enjoy a conference where you feel a major directional change is happening. The Defence Test and Evaluation Conference (DTEC) was such an amazing event. It was the feeling that traditional engineering, V&V and assurance thinking were no longer sufficient. My keynote talk was on connecting many digital twins (from device levels to system…
It was fun joining as a talking head on Auzbiz’s TV interview on AI. After a trite “we shall not worry” answer to the unavoidable “AI becoming sentient” question, I directed the “worry” to the more immediate AI safety and ethical risks. I also joked about how my AI “CoPilot” is writing better code than…
Here is my talk – “Cyber technologies for SME growth – Barriers and Solutions” at CSIRO’s Data61‘s exciting event at the cool Powerhouse Museum.
