Call for Speakers
[un]prompted.au - AI x CyberSecurity: Notes from the Field
18 Sep 2026
19 Sep 2026
ILUMINA Sydney, Australia
unprompted.au
Unprompted.au is an AI-focused security research conference - an Australian edition of the US-based Unprompted conference. Based in Sydney, Unprompted.au sits at the intersection of state-of-the-art AI and Cybersecurity research, with an emphasis on how the field is evolving and where it is heading.
The conference convenes practitioners actively reshaping security workflows in the cyber world where AI is playing a major role - for vulnerability researchers, exploit developers, red/blue teamers, and malware analysts integrating AI to scale analysis, accelerate discovery, and move toward more automated forms of research. Alongside them, researchers and engineers from frontier AI labs provide insight into how these systems are designed, how their capabilities are developing, and the security implications of that trajectory.
The program extends beyond security into core AI research: the techniques, tools, and systems that are redefining what is possible—from automated research workflows to advances in inference, optimisation, and model behaviour.
We intend to continually expand our reach across the broader AI community (touching governance and infrastructure particularly) as the conference grows.
Call for Speakers
29 Apr 2026
31 Jul 2026
Unprompted is about how AI is actually being used today — and where it’s going next.
We are looking for technical, experience-driven talks from people building, breaking, or deploying real systems — as well as those pushing the boundaries of what’s coming:
[AI for Offensive Security]
- Using AI to find, analyse, or exploit vulnerabilities
- Scaling reverse engineering, fuzzing, or bug discovery
- Jailbreaking/prompt injection
- Automation of red-teaming related tasks
- Exploit generation, variant analysis, or automation workflows
- What works, what doesn’t, and where AI falls over
[AI for Defensive Security]
- AI in detection, triage, response, or code review
- Securing AI systems (prompt injection, data exfiltration, abuse)
- Real-world deployments and evaluation of AI-driven defenses
- Failure modes, blind spots, and adversarial behaviour
[AI Research & Tooling (Now and Next)]
- New ideas in agents, reasoning, or autonomy
- Advances in local models, inference, and optimisation (e.g. speculative decoding, quantisation)
- Frontier lab insights into model capabilities and limitations
- Systems that change how practitioners actually use LLMs
- Early-stage research likely to impact security in practice
------------------------------------------------------------------------------------------------------------------------------------
What Makes a Good Talk?
- Technical depth over high-level overviews
- Real-world experience, or clear forward insight
- New ideas, not recycled content
- Honest about failures, not just success stories
If you’re working on AI — whether applying it today or shaping what comes next — and can show what actually happens under the hood, we want to hear from you.
event fee
Travel coverage might be available for selected sessions
If you haven't logged in before, you'll be able to register.
Using social networks to login is faster and simpler, but if you prefer username/password account - use Classic Login.