Back to all courses
AI Security and Generative AI Threats

Cybersecurity and Information Protection

AI Security and Generative AI Threats

5 days · 25 training hours4 scheduled dates
Course code
AD27-CYB-038
Duration
5 days
Training hours
25
Dates and venues
4

Available sessions

Choose a date & venue

4 dates and venues

Please choose a date and venue

Course overview

ALDIYAR Training & Consultancy presents "AI Security and Generative AI Threats", an applied programme on using generative artificial intelligence for clearly defined professional tasks with appropriate human review. Participants examine task framing, instructions, source material, output quality and information protection. The programme distinguishes plausible generated text from verified information and connects productivity opportunities with responsible use, operational integration and clear limits on automation. The learning sequence focuses on ai threat scenarios and trust boundaries, defensive testing and ai security controls, ai use cases and professional task requirements, linking subject knowledge with professional analysis and practical implementation.

Participants work through cases and illustrative information relevant to "AI Security and Generative AI Threats" and cybersecurity and information protection. They practise diagnosing requirements, comparing alternatives, recording assumptions and coordinating decisions with the appropriate stakeholders. Activities use fictional or approved training information and consider resources, information quality, risks and operational constraints. The final exercise brings the concepts together in a workplace action plan. Consequential technical, legal, financial or clinical decisions remain subject to qualified review and the applicable organisational authority.

Learning objectives

  • Explain the principles and requirements relevant to ai threat scenarios and trust boundaries.
  • Analyse the information and assumptions used in defensive testing and ai security controls.
  • Apply appropriate professional methods to ai use cases and professional task requirements.
  • Review controls and responsibilities associated with verification, human oversight and workflow integration.
  • Compare alternatives related to generative ai capabilities, tasks and limitations using explicit criteria.
  • Prepare a workplace action plan with documented roles, resources and review points.

Course Modules

Day 1: AI Threat Scenarios and Trust Boundaries

  • Map users, models, retrieved information and connected tools.
  • Identify untrusted inputs and privileged action boundaries.
  • Review prompt injection, data exposure and misuse scenarios.
  • Assess supplier, configuration and information risks.
  • Prepare threat assumptions and defensive requirements.

Day 2: Defensive Testing and AI Security Controls

  • Test authorised examples of malicious or misleading input.
  • Review output validation and human approval of sensitive actions.
  • Apply least-privilege access and information-handling controls.
  • Monitor misuse, unexpected behaviour and control failures.
  • Prepare remediation and incident escalation responsibilities.

Day 3: AI Use Cases and Professional Task Requirements

  • Define the business task and its current working method.
  • Assess data, information and user requirements.
  • Compare AI assistance with simpler process or analytical alternatives.
  • Identify potential value and consequences of unreliable output.
  • Prepare a bounded use case with acceptance criteria.

Day 4: Verification, Human Oversight and Workflow Integration

  • Review factual correctness, completeness and relevance of outputs.
  • Protect confidential information and reject untrusted instructions.
  • Define qualified human review before consequential decisions.
  • Monitor quality, cost and the need for intervention.
  • Plan a limited pilot and reassessment responsibilities.

Day 5: Generative AI Capabilities, Tasks and Limitations

  • Distinguish generation, prediction, retrieval and ordinary automation.
  • Define suitable tasks and intended professional users.
  • Review common limitations, fabricated content and uncertainty.
  • Compare AI assistance with existing work methods.
  • Prepare a use-case brief and criteria for acceptable output.

Target Audience

  • Cybersecurity and information security professionals.
  • IT operations and system administration personnel.
  • Developers and technology project teams.
  • Risk and compliance specialists reviewing security controls.
  • Authorised security testing and assurance practitioners.
  • Managers responsible for protecting organisational information.

Training methodology

  • Interactive briefings and worked examples on ai threat scenarios and trust boundaries.
  • A facilitated workshop addressing defensive testing and ai security controls.
  • Case analysis of ai use cases and professional task requirements using fictional or approved training information.
  • Role-play or facilitated decision exercises using realistic, non-sensitive workplace scenarios.
  • Peer review of recommendations, documentation and control arrangements with specific feedback.
  • An integrated group exercise developing a practical implementation and review plan for the programme topic.

Training for your team

Need one of these courses for your organisation only, on your dates or at your premises? Our team can plan an in-house program.

Contact our team