For every employee who uses AI assistants at work and needs to know what data is safe to share, which tools to use, and what to do when something goes wrong.
Level
Basic
Study time
6 hours
Lessons
15 in 5 modules
Exam
20 questions
About this course
AI assistants are now built into email, documents, browsers and chat, and most data incidents involving AI start with an ordinary employee trying to get work done faster. This course gives every employee a clear, practical mental model of where prompts and files go, why the type of account matters (consumer, business or enterprise), and how to apply your organization's data classification before you paste, upload or connect anything.
You will practice redacting and substituting realistic examples, learn which kinds of data are red lines (credentials, personal and health data, client confidential material), understand what changes when an assistant can read your mailbox or act on your behalf, and recognize hidden instructions planted in emails and web pages. The course closes with what to do in the first hour after a mistake and a personal checklist you can keep. Product features are described as of October 2026 and in general terms; always follow your own organization's approved-tool list and policy. The course is educational and is not legal advice. The Wirewalk Institute certificate attests that you completed the course and passed its automated multiple-choice exam; it does not certify workplace competence, and it is not endorsed by OWASP, NIST or any vendor named here. To report an error, contact the Wirewalk Institute through institute.wirewalk.com. The course is reviewed every six months, and sooner when a cited vendor page or law changes.
What you will be able to do
Explain what can happen to prompts, uploads and conversation history in consumer, business and enterprise AI services
Classify information against an organization's data classes before sharing it with an AI tool
Redact or substitute sensitive details so a task can still be completed safely
Identify the extra risk introduced by connected apps, file access and AI agents acting on your behalf
Recognize hidden instructions in emails, documents and web pages that try to manipulate an AI assistant
Verify AI output before sharing it or acting on it
Report and help contain a suspected AI-related data exposure promptly
Syllabus
5 modules · 15 lessons · about 6 hours
1
Module 1: Where Your Data Goes
What actually happens to a prompt or upload, and why the account and tool you use matter.3 lessons
What Happens to a Prompt: Retention, Training Use and Human Review
Consumer, Business and Enterprise Accounts: Why the Tier Changes the Rules
Approved Tools Versus Shadow AI
2
Module 2: Classify Before You Paste
Applying data classes to AI use and making tasks safe through redaction and substitution.3 lessons
Applying Your Organization's Data Classes to AI Use
Red Lines: Credentials, Personal, Health, Financial and Client Data
Redacting, Generalizing and Using Synthetic Examples
3
Module 3: Files, Connected Apps and Agents
How uploads, mailbox and drive connectors, and agents that act for you change the risk.3 lessons
Uploading Documents, Spreadsheets and Screenshots Safely
Connectors to Email, Drive and Calendar: What the Assistant Can Now See
When an AI Agent Acts for You: Permissions and Confirmations
4
Module 4: Manipulation and Unreliable Output
Hidden instructions aimed at your assistant and checking output before it leaves your hands.3 lessons
Hidden Instructions in Emails and Web Pages: Prompt Injection for Everyone
Checking AI Output Before You Share or Act on It
Sharing AI Chats, Links and Generated Files
5
Module 5: When Something Goes Wrong
Recognizing an exposure, the first hour of response, and habits that prevent the next one.3 lessons
Recognizing an AI-Related Data Exposure
Reporting and Containing: The First Hour
Your Personal Safe-AI Checklist
Exam and certificate
The exam unlocks once you have completed every lesson. It draws 20 questions from a larger bank, spread across the modules, with the answer options shuffled for each attempt. You have 30 minutes and need 80% to pass. After a failed attempt you can try again after 24 hours, up to 3 attempts in total. Feedback is given per module, never as the answer key.
Passing earns a certificate of completion and assessment from the Wirewalk Institute of AI with a unique ID that anyone can check at institute.wirewalk.com/verify. It is not an accredited qualification or a vendor certification. How certification works.
The Wirewalk Institute of AI is independent. This course teaches publicly documented products and is not affiliated with or endorsed by OpenAI ChatGPT, Microsoft Copilot, Google Gemini, or Anthropic Claude. Product and company names are trademarks of their respective owners.
Follow one fictional US distributor through four everyday AI mistakes (a guarantor's credit file uploaded to a personal account, a customer letter promising a fee waiver that does not exist, a vendor email with hidden instructions for the assistant, and a note-taker that sent a customer the internal margin floor) and practice the decisions that prevent or contain each one in three short Decision Labs.
A practitioner course on securing applications that use large language models, organized around the 2026 OWASP Top 10 for LLM Applications. Every risk comes with a hypothetical scenario, the root cause, and concrete defenses in code and configuration.
A deep course on prompt injection: why it exists, how direct, indirect and multimodal payloads turn into data exfiltration and goal hijack in agents, why filters alone fail against adaptive attackers, and which architectural patterns and tests actually reduce risk.