Wirewalk Instituteof AI
sec-001 Free intro Available

Using AI Safely at Work: The Essentials

A free course of about 75 minutes for anyone who uses AI assistants at work: decide what is safe to share, check what comes back, spot hidden instructions, and act fast when something goes wrong.

Level
Free intro
Study time
1 hours
Lessons
7 in 3 modules
Exam
10 questions

About this course

AI assistants are now part of everyday work, and most problems with them start with an ordinary person trying to get a job done quickly. This free course follows Prairie Line Supply Co., a fictional composite of a 640-person US building-supplies distributor, through its first months with approved AI tools: a credit analyst who uploads a guarantor's credit application to his personal account, a customer letter that promises a fee waiver the company does not offer, a vendor email with hidden instructions aimed at the assistant reading it, and a free AI note-taker that sends a customer the branch's internal margin floor.

In each case you see the actual prompt, email or consent screen, what the AI produced, what an expert would have checked, and what the company changed, with the cost of each fix in hours and the problems it left unsolved. You practice the same decisions on short exercises with model answers and in three four-minute Decision Labs (one story in three acts) where you play the credit manager, the branch office manager and the branch manager: which account and settings make a use acceptable, what is safe to paste and how to rewrite a request so it still works, how much to check AI output given the stakes, how to respond when an assistant has read injected instructions, when to use the approved tool or ask for approval, and what to do in the first hour after a mistake. Real-world examples (court orders, published vulnerability research, FBI crime data, Microsoft and OpenAI documentation, NIST and OWASP guidance) are cited with dates. The core path takes about 75 minutes: seven short lessons with exercises, the three labs and a 10-question exam. You finish with a one-page AI Safety Card for your own job, scored against a published rubric, with guidance for people whose employer has no approved tool yet. Product details are described as of October 2026; always follow your own organization's policy and approved-tool list. The Wirewalk Institute certificate attests that you completed the course and passed its automated exam; it is not endorsed by OWASP, NIST or any vendor named here.

What you will be able to do

  • Determine whether an AI tool's account type and settings make a given work use acceptable, and name what to check first
  • Classify the parts of a work item as green, amber or red and rewrite the request so the AI can help with no red item left
  • Choose a level of checking that matches the stakes and locate unsupported claims in an AI-drafted message
  • Identify injected instructions in content an assistant has read and choose the safe response, including verification through a channel the requester does not control
  • Decide between the approved tool, an approval request or no AI for a tool, extension or note-taker, based on what it can reach and who controls the data
  • Carry out the first hour after an AI-related mistake: stop, contain without destroying evidence, report the five facts and change exposed secrets

Syllabus

3 modules · 7 lessons · about 1 hours

Module 1: What You Share and Where It Goes

How the account a tool is signed in with decides what happens to your data, and a fast test for what is safe to paste.
2 lessons
  1. Where Your Prompt Goes: Personal Accounts, Business Accounts and Settings
  2. The Traffic-Light Test: Deciding What Is Safe to Paste
  • Decision Lab · 4 minThe Guarantor FileA credit analyst uploaded a guarantor's credit application to his personal ChatGPT account two days ago. Decide what happens in the next hour, then what rule your team lives with.

Module 2: Output You Check and Instructions You Ignore

Checking AI output in proportion to the stakes, and recognizing hidden instructions in what an assistant reads.
2 lessons
  1. Confident and Wrong: Checking AI Output in Proportion to the Stakes
  2. Hidden Instructions: Prompt Injection in Email, Files and Web Pages
  • Decision Lab · 4 minThe Wichita InboxAn AI-drafted letter promises a fee waiver that does not exist, and ten days later your assistant tells you a supplier has changed banks. Decide what you check, whom you call and what you ask IT to change.

Module 3: Approved Tools and Fast Reporting

Why approved tools beat shadow AI, including note-takers and extensions, the first hour after a mistake, and your own AI Safety Card.
3 lessons
  1. Approved Tools Versus Shadow AI: Extensions, Note-Takers and Personal Accounts
  2. The First Hour After a Mistake: Stop, Contain, Report
  3. Capstone: Build Your Own AI Safety Card
  • Decision Lab · 4 minThe Note-TakerA free AI note-taker has just emailed your internal margin floor to a customer. Decide what you do in the first hour, what you say when the customer calls, and what your branch uses for meetings from now on.

Exam and certificate

The exam unlocks once you have completed every lesson and played each Decision Lab to an ending. It draws 10 questions from a larger bank, spread across the modules, with the answer options shuffled for each attempt. You have 20 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, Anthropic Claude, or Microsoft (Copilot Chat, Teams, Entra ID). Product and company names are trademarks of their respective owners.

More in AI Security: Securing AI Systems

All 13 courses in this track
sec-101 Basic Available

Using AI Safely: Data Handling for Every Employee

A plain-language course on handling data safely with AI tools: what happens to what you type or upload, how to classify before you paste, how connected apps and agents change the picture, and how to report a mistake quickly.

6 h · 15 lessons See price at checkout
sec-201 Intermediate Available

AI Security Fundamentals: The OWASP Top 10 for LLMs

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.

14 h · 18 lessons See price at checkout
sec-210 Intermediate Available

Prompt Injection and Agent Security

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.

12 h · 17 lessons See price at checkout