Generative AI Security (OWASP LLM Top 10) Fundamentals Training
A one-day fundamentals programme on OWASP LLM Top 10. Learn how to reduce risks from AI chatbots leaking data, following malicious instructions or mishandling their outputs. Participants work through practical examples and a guided case before drafting actions for their own organisation.
Who should attend: Application owners, developers, IT, security, data and technical risk staff. No prior framework knowledge is required.
HRD Corp claim support available
Programme Agenda
9:00 AM to 10:00 AM
LLM Applications and the Current Top 10
Introduce the OWASP LLM Top 10 2026 and draw a simple chatbot architecture. Identify models, prompts, retrieval sources, tools and user interfaces without treating the model as the whole application.
10:00 AM to 10:15 AM
Break
10:15 AM to 11:30 AM
Untrusted Instructions and Excessive Agency
Explain prompt injection and risks from giving an assistant too much authority. Separate untrusted content from trusted policy and identify decisions needing enforcement outside the model.
11:30 AM to 12:30 PM
Sensitive Data and Knowledge Sources
Explore inappropriate disclosure, over-broad retrieval access, unsafe data ingestion and misleading source content. Match safeguards to permissions, provenance and the intended use of the information.
12:30 PM to 1:30 PM
Lunch
1:30 PM to 2:45 PM
Outputs and Downstream Systems
Consider unsafe output handling, unreliable claims and actions based on generated content. Introduce validation, safe display, constrained integrations and human review for consequential decisions.
2:45 PM to 3:00 PM
Break
3:00 PM to 4:00 PM
Operational and Supplier Risks
Recognise dependency, model supply chain and uncontrolled consumption exposure. Establish evaluation, rate or cost limits, monitoring and change review as parts of application operations.
4:00 PM to 5:00 PM
Practical Exercise: A Document Chatbot
Review a fictional retrieval assistant and illustrative unsafe responses. Map risks to controls, identify verification evidence and draft five remediation priorities and remaining limitations.
Key Outcomes
- Explain common LLM application risks
- Identify untrusted input and data access boundaries
- Choose output and integration safeguards
- Draft basic evaluation and remediation actions
Training Mode Physical / Online / Hybrid
HRD Corp Subject to programme eligibility and HRD Corp approval
Level Fundamentals / Beginner. No prior framework knowledge required.
Duration 1 Day | 9:00 AM to 5:00 PM (6.5 learning hours, excluding breaks)
Venue In-house at the client's premises, or delivered via the client's preferred platform (Microsoft Teams, Zoom, or equivalent)
Assessment Group case exercise and a 10-question knowledge check
Certificate Orbix Certificate of Completion upon full attendance; not a professional certification
Malaysia Relevance
High relevance for Malaysian organisations deploying generative AI applications. Examples distinguish ordinary staff use from building an application with retrieval, integrations and delegated actions.
Practical Takeaways
A generative AI application risk map and security evaluation checklist.
Download a sample practice worksheet (TXT). This fictional exercise is a preview of the learning approach. About Orbix; contact us to confirm the facilitator and relevant experience for your delivery.
Course Completion and Next Steps
Participants receive an Orbix Certificate of Completion upon full attendance. This introductory course does not confer professional certification or establish an organisation's compliance. Further implementation and assessment depend on the organisation's scope and obligations.
Frequently Asked Questions
Related Fundamentals Training
AI Agent Security (OWASP Agentic Top 10) Fundamentals AI Agent Runtime Security (OWASP Agent Control Standard) Fundamentals AI Security Verification (OWASP AISVS) Fundamentals AI Security Controls (CSA AICM) Fundamentals View all fundamentals programmes
Reference material: OWASP: GenAI LLM Top 10 2026.