Offensive Security and Adversarial AI Training

This program prepares practitioners and leaders to understand, anticipate, and defend against the offensive techniques now emerging at the intersection of cybersecurity and artifi cial intelligence. Drawing on Dr. Morrison’s formal approval under Anthropic’s Cyber Verifi cation Program, which authorizes qualifi ed dual-use research and education in offensive security contexts, this track explores the tools, methods, and ethical guardrails that defi ne responsible adversarial work.

Topics include:

  • Foundations of offensive security and the modern red team discipline
  • AI-augmented reconnaissance, exploitation, and post-exploitation workfl ows
  • Prompt injection, jailbreak, and model manipulation techniques, with their defensive counterparts
  • Adversarial machine learning, including evasion, poisoning, and model extraction
  • Governance, disclosure, and the ethics of offensive research in regulated environments
  • Practical laboratory exercises using sanctioned environments and controlled datasets

The program is designed for security professionals, red and purple team members, graduate students, and leaders responsible for authorizing or overseeing offensive security programs. Engagements can be scoped as short executive briefi ngs, multi-day intensive workshops, or semester-length academic courses.

Education And Training

Cybersecurity and artificial intelligence are moving faster than most organizations can absorb. Education is how leaders close that gap. At PKM Cyber Consultants, we approach teaching as a form of stewardship. The people making security decisions, writing the policies, or operating the tooling deserve instruction grounded in both academic rigor and practical experience.

Dr. Paul Morrison brings together more than fifteen years of executive cybersecurity leadership, doctoral research in computer science, and active teaching at the university level. His instruction draws from his work as a Business Information Security Officer at a global financial institution, his ongoing research in AI system design, and his forthcoming role as Lecturer in Cyber Security at De Montfort University Cambodia. Each program is shaped to meet learners where they are, whether they are stepping into the field for the first time or leading a team of hundreds.

University and Graduate Instruction

Dr. Morrison teaches at the university level in both undergraduate and graduate programs, covering foundational cybersecurity, advanced threat intelligence, artificial intelligence in security operations, and the governance frameworks that increasingly shape modern practice. Current engagements include his upcoming lectureship at De Montfort University Cambodia beginning September 2026. Guest lectures, curriculum development, and program advisory work are also available for institutions seeking a practitioner voice in the classroom.

Executive and Professional Development

For organizations, PKM Cyber Consultants delivers tailored programs that translate complex security and AI topics into language that boards, executives, and technical teams can act on. Typical engagements include:

  • Cybersecurity strategy briefi ngs for boards and C-suite audiences
  • AI governance and responsible AI workshops for leadership teams
  • Incident response tabletop exercises and scenario-based learning
  • Role-specifi c training for security teams, risk officers, and audit functions

 
Programs are delivered in person, virtually, or in hybrid formats, and are built to fi t the operational rhythm of the host organization.

Offensive Security and Adversarial AI Training

This program prepares practitioners and leaders to understand, anticipate, and defend against the offensive techniques now emerging at the intersection of cybersecurity and artifi cial intelligence. Drawing on Dr. Morrison’s formal approval under Anthropic’s Cyber Verifi cation Program, which authorizes qualifi ed dual-use research and education in offensive security contexts, this track explores the tools, methods, and ethical guardrails that defi ne responsible adversarial work.

Topics include:
  • Foundations of offensive security and the modern red team discipline
  • AI-augmented reconnaissance, exploitation, and post-exploitation workfl ows
  • Prompt injection, jailbreak, and model manipulation techniques, with their defensive counterparts
  • Adversarial machine learning, including evasion, poisoning, and model extraction
  • Governance, disclosure, and the ethics of offensive research in regulated environments
  • Practical laboratory exercises using sanctioned environments and controlled datasets

The program is designed for security professionals, red and purple team members, graduate students, and leaders responsible for authorizing or overseeing offensive security programs. Engagements can be scoped as short executive briefi ngs, multi-day intensive workshops, or semester-length academic courses.

Offensive Security and Adversarial AI Training

This program prepares practitioners and leaders to understand, anticipate, and defend against the offensive techniques now emerging at the intersection of cybersecurity and artifi cial intelligence. Drawing on Dr. Morrison’s formal approval under Anthropic’s Cyber Verifi cation Program, which authorizes qualifi ed dual-use research and education in offensive security contexts, this track explores the tools, methods, and ethical guardrails that defi ne responsible adversarial work.

Topics include:

  • Foundations of offensive security and the modern red team discipline
  • AI-augmented reconnaissance, exploitation, and post-exploitation workfl ows
  • Prompt injection, jailbreak, and model manipulation techniques, with their defensive counterparts
  • Adversarial machine learning, including evasion, poisoning, and model extraction
  • Governance, disclosure, and the ethics of offensive research in regulated environments
  • Practical laboratory exercises using sanctioned environments and controlled datasets

The program is designed for security professionals, red and purple team members, graduate students, and leaders responsible for authorizing or overseeing offensive security programs. Engagements can be scoped as short executive briefi ngs, multi-day intensive workshops, or semester-length academic courses.

Engage

To inquire about availability, curriculum customization, or speaking engagements, please visit the Contact page or reach out directly through the inquiry form.

Program One

Red Team Fundamentals

3-Day program

Core offensive security concepts, methodology, and tooling for defensive teams building internal red team capability. Participants complete hands-on laboratory exercises covering reconnaissance, initial access, and post-exploitation, and leave with a working framework for future engagement design.

Program Two

Adversarial AI Defense Workshop

2-Day program

How attackers target machine learning systems and the defensive patterns that actually work. Covers prompt injection, data poisoning, model extraction, and evasion attacks, with practical defense planning for AI systems currently in production.

Program Three

AI-Augmented SOC Operations

2-Day program

How security operations centers can responsibly integrate AI assistants into their daily workflows. Addresses prompt engineering for SOC tasks, operational security when using AI, and the boundary between augmentation and automation in detection and response.

Program Four

Tabletop Exercise Facilitation

1-Day program

Designing and running executive tabletop exercises for cyber crisis preparedness. Participants leave with a complete tabletop facilitation template and the skills to run engagements for their own leadership teams

Program Five

Custom Curriculum Development

2-Day program

Variable, available for retained engagements Description: Bespoke multi-week programs tailored to specific organizational needs. Ideal for organizations seeking to build systematic cybersecurity capability across their technical teams. Engagement scope determined through discovery consultation.