AI Security Speakers: Experts in AI Safety, Risk & Defense

In this resource we present who we consider to be the leading AI Security speakers of 2026.

We list a ton of information of cyber security conferences and we post a lot of cutting-edge AI webinars.

* If you'd like to add your name or company to this list - get in contact!

Discover the latest bleeding-edge AI Security Demonstrations

Core AI Safety & Alignment

This group includes researchers and thinkers focused on the fundamental question of how advanced AI systems should behave as they scale.

Their talks tend to cover alignment, interpretability, model objectives, and long-term safety risks—topics that influence how “secure-by-design” AI is framed across industry and policy.

For AI Security audiences, these speakers help connect technical failure modes to real-world impact, shaping how teams think about safety requirements, evaluation, and assurance long before deployment decisions are made.

Core AI Safety & Alignment
Paul Christiano	Alignment Research Center / Former OpenAI	Dwarkesh Podcast Live Events; AI Alignment Public Talks
Sheila McIlraith	University of Toronto / Vector Institute	AAAI 2023; IJCAI 2016
Jeff Clune	University of British Columbia / Former OpenAI	NeurIPS Workshops; AI Safety & Interpretability Workshops
David Duvenaud	University of Toronto / Vector Institute	NeurIPS Workshops; ICLR Workshops
Gillian Hadfield	University of Toronto / Vector Institute	Stanford AI100 Meetings; AI Governance Conferences
Roger Grosse	University of Toronto	NeurIPS Workshops; ICLR Workshops
Nicolas Papernot	Google Brain / Vector Institute	NeurIPS 2022 Workshops; DPFM@ICLR 2024

AI Security, Robustness, and Privacy

These speakers sit closer to the engineering edge of AI risk: robustness under attack, privacy leakage, adversarial manipulation, and how real systems fail in production.

Their sessions often translate abstract safety concerns into concrete controls—testing, hardening, monitoring, and governance—across models and pipelines.

If you’re building or buying AI security tools, this is the category that most directly informs practical defenses, from threat modeling and red teaming to privacy-preserving design and resilient deployment patterns.

AI Security, Robustness & Privacy
Luka Ivezic	Information Security Forum (ISF)	ISF Secure & Trusted AI Events
Marin Ivezic	Applied Quantum	AI & Critical Infrastructure Security Conferences
Sheila McIlraith	University of Toronto / Vector Institute	ICAPS 2023; AAAI 2023
Jeff Clune	University of British Columbia	NeurIPS; AI Safety Workshops
David Duvenaud	University of Toronto / Vector Institute	UK–Canada Frontiers of Science: AI
Nicolas Papernot	Google Brain / Vector Institute	IEEE SaTML; UK–Canada Frontiers of Science: AI

AI Governance and Systemic Safety

AI governance speakers focus on how organisations and governments can manage AI risk at scale—through policy, standards, accountability, and oversight.

Expect themes like compliance, auditability, model assurance, procurement controls, and systemic safety frameworks that influence how AI is approved and operated. For security leaders, these talks are useful because they bridge the gap between technical risk and organisational decision-making, helping teams define responsibilities, control objectives, and measurable requirements for safe AI adoption.

AI Governance & Systemic Safety
Gillian Hadfield	University of Toronto / Vector Institute	AI100 Governance Events; AI Policy Workshops
David Duvenaud	University of Toronto	AI Frontiers of Science; AGI Governance Meetings
Sheila McIlraith	University of Toronto	KR 2014; AAAI 2023
Jeff Clune	University of British Columbia	Frontier AI Policy Panels; Safety Workshops
Nicolas Papernot	Google Brain / Vector Institute	IEEE Secure & Trustworthy ML (SaTML)

AI-Powered Cybersecurity Leaders

This table highlights executives and operators driving the commercial side of AI-driven defense—how AI is being used to improve detection, response, exposure management, and security operations.

Their talks are typically grounded in outcomes: faster triage, better signal-to-noise, improved resilience, and practical deployment lessons in large environments.

For event audiences, this category offers a view into where budgets are moving, what buyers are prioritising, and how vendors are positioning AI capabilities across cloud, endpoint, and SOC workflows.

AI-Powered Cybersecurity Leaders
Beenu Arora	Cyble	NASSCOM US CEO Forum 2025
John D. Loveland	StrikeReady	AI-Driven SOC & Cyber Operations Conferences
Yevgeny Dibrov	Armis	OT & AI-Powered Cyber Exposure Conferences
Matthew Prince	Cloudflare	Cloudflare Connect; Global Security Conferences
Jay Chaudhry	Zscaler	Zenith Live 2021; Zenith Live 2025
Allie Mellen	Forrester Research	RSA Conference; Black Hat; HOPE

Cybersecurity & AI Safety Ecosystem Shapers

These ecosystem shapers sit at the intersection of security practice, community education, and forward-looking risk narratives.

These conference speakers often speak on how architectures are evolving, how automation changes operations, and how organisations should adapt programs for modern threats—including AI-enabled abuse.

Their value is breadth: they connect trends across vendors, practitioner communities, and emerging research, making them ideal for conferences that want practical guidance with strategic perspective. Expect insights on culture, governance, and scalable security design.

Cybersecurity & AI Safety Ecosystem Shapers
Andy Ellis	YL Ventures / Former Akamai	RSA Conference; Black Hat
Allison Miller	Independent Security Researcher	Fraud & Abuse Prevention Conferences
Bob Rudis	Rapid7	Data-Driven Security & SOC Analytics Events
Chris Wysopal	Veracode	RSA Conference; OWASP Events
Jim Tiller	Cynomi	CISO & Cybersecurity Strategy Conferences
Toryn Klassen	Vector Institute	AI Safety Reading Group Events
Michael Zhang	Vector Institute	AI Trust & Safety Workshops

Wrapping Up

The landscape of AI safety and security spans a remarkable breadth of expertise—from foundational alignment research to hands-on defensive engineering, governance frameworks, and commercial innovation. What connects these speakers is a shared recognition that securing AI systems requires collaboration across disciplines.

Alignment researchers help us understand why models might fail; robustness experts show us how to harden them; governance leaders define what controls organizations need; and industry practitioners demonstrate where these concepts translate into operational reality.

For security professionals planning events or building programs, this diversity isn't a complication—it's an asset. The most effective AI security strategies draw from all these perspectives, bridging theoretical risk with practical defense and organizational accountability.