Conference Description
Key Takeaways
- AMLUCS is the UK’s only government-backed conference dedicated to the intersection of artificial intelligence and cybersecurity
- The fourth annual event focuses on practical, real-world applications rather than theoretical or commercial presentations
- Core themes include AI security threats, offensive and defensive AI operations, multi-agent system architectures, governance frameworks and explainability
- Aimed at security engineers, data scientists, researchers, policymakers and operational practitioners
- Features a pre-conference training course on Validated AI Red Teaming for Production AI Systems
Introduction
AMLUCS returns for its fourth year as the United Kingdom’s sole government-backed conference examining the convergence of artificial intelligence and cybersecurity. The event serves security engineers, data scientists, researchers, vendors, policymakers and operational practitioners who work directly with AI-enabled systems in defensive and offensive contexts. As organisations increasingly deploy machine learning models across critical infrastructure and security operations, the need for rigorous, practitioner-focused discourse on vulnerabilities, mitigations and governance has become acute. AMLUCS addresses this gap by prioritising applied research and deployment experience over commercial messaging.
About AMLUCS
AMLUCS operates as a not-for-profit, practitioner-led initiative designed to bridge the divide between academic research and operational security practice. The conference benefits from the support of an international research community and a dedicated Technical Committee responsible for curating the programme content. This structure ensures that sessions reflect genuine technical depth rather than vendor-driven narratives.
The event’s government backing distinguishes it within the UK conference landscape, providing a platform where public sector priorities intersect with private sector innovation. By maintaining its not-for-profit status, AMLUCS can focus on strengthening collaboration across the broader UK cyber and AI community without commercial pressures influencing the agenda.
Core Conference Themes
The conference programme is organised around five interconnected themes that reflect the current state of AI security research and practice. These themes acknowledge that securing AI systems requires expertise spanning technical implementation, architectural design, governance frameworks and ethical considerations.
AI Security Threats, Vulnerabilities and Mitigations
Machine learning models present novel attack surfaces that traditional security frameworks were not designed to address. Adversarial inputs, data poisoning, model extraction and inference attacks represent categories of threat that require specialised detection and mitigation strategies. This theme examines practical approaches to identifying and addressing these vulnerabilities in production environments.
Offensive and Defensive AI in Cyber Operations
The application of AI to both attack and defence represents one of the most significant shifts in cybersecurity operations. Autonomous agents capable of reconnaissance, exploitation and lateral movement are no longer theoretical concerns. Equally, defensive systems leveraging machine learning for threat detection, anomaly identification and automated response are becoming standard components of security operations centres. Understanding the capabilities and limitations of both applications is essential for practitioners operating in this space.
Multi-Agent Systems and Secure Architecture Design
As organisations deploy increasingly complex AI systems involving multiple interacting agents, architectural security becomes paramount. The design decisions made during system development have profound implications for resilience, auditability and failure modes. This theme addresses the challenges of building secure multi-agent architectures that can operate reliably in adversarial environments.
Governance, Risk Assessment and Emerging Standards
Regulatory frameworks governing AI deployment are evolving rapidly across jurisdictions. Organisations must navigate emerging compliance requirements while maintaining operational effectiveness. Risk assessment methodologies specific to AI systems remain an area of active development, with practitioners seeking frameworks that can accommodate the unique characteristics of machine learning models, including their opacity and potential for unexpected behaviours.
Explainability, Interpretability and Fairness
Security applications of AI raise particular concerns around explainability and fairness. When machine learning models inform decisions about threat classification, access control or incident response, the ability to understand and justify those decisions becomes operationally and legally significant. This theme explores techniques for achieving meaningful interpretability without sacrificing model performance.
Programme Structure
The conference features a curated programme combining multiple session formats. Technical talks draw on applied research and real-world deployment experience, providing attendees with actionable insights rather than abstract theory. Keynote sessions feature international experts alongside UK government leaders, offering perspectives on both technical developments and policy direction.
Practitioner panels address current challenges and emerging risks through structured discussion, allowing attendees to hear diverse viewpoints on contested or evolving topics. The format encourages engagement with the complexities and trade-offs inherent in securing AI systems at scale.
A two-day pre-conference training course on Validated AI Red Teaming for Production AI Systems provides intensive, hands-on preparation for practitioners seeking to develop or enhance their AI security testing capabilities. Red teaming has emerged as a critical discipline for identifying vulnerabilities in deployed AI systems before adversaries can exploit them.
Who Should Attend
AMLUCS is designed for professionals actively engaged in building, testing, defending or governing AI-enabled systems. The conference serves those working at the operational level rather than those seeking introductory material. Relevant roles include security engineers responsible for protecting AI infrastructure, data scientists developing models for security applications, researchers investigating AI vulnerabilities and mitigations, policymakers shaping regulatory approaches, and vendors developing security products incorporating machine learning.
The emphasis on practical application means attendees should expect content grounded in deployment realities rather than speculative futures. Those seeking to understand what currently works in production environments, and what challenges remain unsolved, will find the programme particularly valuable.
Industry Context
The rapid integration of AI into security operations has created an urgent need for forums where practitioners can share knowledge without commercial interference. Traditional cybersecurity conferences often struggle to address AI-specific concerns with sufficient technical depth, while AI conferences frequently overlook security implications. AMLUCS occupies this intersection, providing a venue where the unique challenges of securing machine learning systems receive focused attention from those with direct operational experience.
The UK’s position as a significant player in both AI research and cybersecurity makes a dedicated national conference particularly valuable. Government backing signals recognition that the security of AI systems represents a strategic priority requiring sustained investment in community development and knowledge sharing.