Track 1 · AI Training & Certification

AI Foundations

A practical learning track for business professionals who want to use AI tools with confidence, safety, and genuine understanding — not just hype.

6 Modules
~2h Total time
Cert On completion
0 / 6
01
What AI Actually Is (And Isn't)
~20 min
02
Working With AI: The Skill That Matters Most
~20 min
03
Critical Thinking With AI
~20 min
04
AI in Your Workflow
~20 min
05
Data, Privacy & Security
~15 min
06
Building Your AI Practice
~15 min
Module 01

What AI Actually Is
(And Isn't)

The problem this solves

Most professionals operate with a mental model of AI shaped by marketing hype and science fiction. This creates either over-trust or unnecessary fear — both of which lead to poor decisions at work.

Pattern Recognition, Not Thinking

AI systems analyse massive datasets to identify patterns and generate predictions. When you interact with a chatbot, it's predicting the most likely next word — not "understanding" your question. This tells you where AI excels (pattern-heavy tasks) and where it doesn't (novel judgment calls).

Three Types You'll Encounter

Generative AI creates new content — text, images, code. Predictive AI forecasts outcomes — demand planning, risk scoring. Automation AI follows rules to handle repetitive tasks. Most workplace tools blend all three.

Why Now? What Actually Changed

AI concepts have existed for decades. The explosion isn't a sudden intelligence breakthrough — it's three things converging: dramatically cheaper computing power, oceans of available training data, and interface design that finally makes AI accessible to non-engineers.

The Confidence Problem

AI presents everything with equal confidence — correct answers, reasonable guesses, and complete fabrications all arrive in the same calm, authoritative tone. This isn't deception; it's a structural feature. Understanding this will save you from more mistakes than any other insight in this course.

Practical Exercise

Identify the AI at Work

Three real workplace scenarios. For each one: what type of AI is at play, what is it actually doing, and where does the human judgment still sit? Click each to reveal the analysis.

Scenario A

Your email client suggests three possible replies to a client message asking about project timelines.

Tap to reveal →
Generative AI. It's predicting plausible responses based on email context and common patterns. Your judgment: does the suggested tone match this client? Are the timelines actually correct?
Scenario B

Your CRM flags that three accounts have a high probability of churning in the next quarter.

Tap to reveal →
Predictive AI. Analysing engagement patterns, support tickets, and usage data to identify risk. Your judgment: do you know something the data doesn't? The model flags risk — you choose the response.
Scenario C

An AP system automatically matches incoming invoices to purchase orders and routes exceptions for review.

Tap to reveal →
Automation AI with predictive elements. Rule-based matching handles routine; ML flags edge cases. Your judgment: reviewing exceptions, catching errors, handling invoices that don't fit the pattern.
Takeaway

AI is a powerful pattern-matching and generation tool. Understanding what it is makes you better at using it — and better at spotting when it's the wrong tool for the job.

Module 02

Working With AI:
The Skill That Matters Most

The problem this solves

Most people interact with AI the way they'd type a Google search — short, vague, and hoping for the best. The quality gap between a poorly prompted and well prompted interaction is enormous, and most professionals don't know it exists.

Prompt craft is a communication skill

The same principles that make you good at writing a brief, delegating to a team member, or scoping a project make you effective with AI. Think of every interaction as a delegation: you're handing a task to a highly capable but context-blind assistant.

  • Context — tell it who you are and what you're working on
  • Instruction — be specific about what you want it to do
  • Constraints — set boundaries: length, tone, audience, format
  • Format — specify how you want the output structured
Before & After — The Same Task, Two Approaches Live Example
✗ Vague prompt
"Write me an email about the project update"
Typical output
A generic, bland email that could be about any project, addressed to no one in particular, with placeholder information. You'll spend more time editing it than you saved.
✓ Structured prompt
"Draft a project status email to my client Sarah (CFO, formal but warm tone). Project: CRM migration. Key updates: Phase 1 complete, 2 days early. Phase 2 training starts Monday. One risk: API integration pending vendor response — expected Thursday. Under 200 words. End with a clear next step."
Typical output
A specific, ready-to-send email with the right tone, accurate details, appropriate structure, and a clear call to action. Minor tweaks at most.

The Iterative Approach

Don't treat AI as a vending machine. Treat it as a conversation. Your first prompt gets a first draft. Follow-ups refine it: "Make the tone more direct." "Add budget implications." "Restructure with the conclusion first." This is where the real value sits.

Six Patterns for Business

Summarise (reduce length, keep meaning), Analyse (find patterns, draw conclusions), Draft (create from a brief), Brainstorm (options with constraints), Compare (evaluate alternatives), Extract (pull data from unstructured content).

Practical Exercise

The Three-Prompt Challenge

Pick a real task from your workday. Write three progressively better prompts for it. Compare the outputs at each level to see the quality difference firsthand.

  1. 01Write the prompt you'd normally use — the quick, instinctive version
  2. 02Rewrite it using the Context + Instruction + Constraints + Format framework
  3. 03Take the output from Prompt 2 and write a follow-up that refines the weakest part
Takeaway

The quality of AI output is directly proportional to the quality of your input. This is a learnable, practicable skill — not a talent.

Module 03

Critical Thinking
With AI

The problem this solves

The two failure modes are equally dangerous — blindly trusting every AI output and refusing to use AI at all. Professionals need a practical framework for evaluating what AI gives them, calibrated to the stakes involved.

Why AI sounds right even when it's wrong

Language models generate text by predicting the most likely next word. This means they're optimised for plausibility, not accuracy. A fabricated statistic and a real one are produced by the same mechanism — which is why AI never hedges or says "I'm making this up."

  • Factual fabrication — invented statistics, false citations, non-existent sources
  • Outdated information — training data has a cutoff; "current" facts may not be
  • Plausible-but-wrong reasoning — logical structure that reaches a flawed conclusion
  • Subtle framing bias — one perspective presented as balanced or complete
  • Missing context — omitting crucial caveats, exceptions, or nuances
Low stakes
Light Check
Brainstorming, first drafts, internal notes — skim for obvious issues
Medium stakes
Verify Claims
Client content, reports, presentations — fact-check key data points and sources
High stakes
Full Review
Financial, legal, compliance, public-facing — independent verification of all claims

The Verification Framework

Match your checking effort to the stakes. A brainstorming list for an internal meeting needs a quick scan. A client-facing report needs every number confirmed. Legal or compliance content needs expert review regardless of how confident the output looks.

Domain Expertise Matters More Now

AI is most useful when you already know enough to evaluate its output. The less you know about a topic, the harder it is to spot errors. AI amplifies expertise rather than replacing it — be most cautious outside your area of knowledge.

Practical Exercise

Spot the Problem

Three AI-generated outputs. One is accurate, one contains a subtle error, one is a confident hallucination. Click each to reveal the analysis.

Output A

"Australia's Productivity Commission was established in 1998, replacing the former Industry Commission, and reports directly to the Treasurer."

Tap to analyse →
Accurate. The Productivity Commission was indeed established in 1998 and reports to the Treasurer. A quick check of the official .gov.au site confirms this.
Output B

"According to a 2024 McKinsey report, 47% of Australian SMEs have fully integrated AI into their core operations, up from 12% in 2022."

Tap to analyse →
Likely hallucinated. The specificity (47%, McKinsey, 2024) creates false authority. Does this report exist? A 35-point jump in two years for "full integration" should trigger immediate scepticism.
Output C

"Under Australian consumer law, businesses with fewer than 20 employees are classified as small businesses under the unfair contract terms provisions."

Tap to analyse →
Subtle error. The employee threshold was changed by legislation. AI may cite an outdated figure. Legal details demand independent verification — never rely on AI for statutory thresholds without checking.
Takeaway

AI is a first-draft machine, not a final-answer machine. Your judgment is the quality control layer, and that layer is non-negotiable.

Module 04

AI in Your Workflow

The problem this solves

People either try to use AI for everything (and get frustrated) or use it for almost nothing (and miss real value). The gap isn't enthusiasm — it's a systematic way to audit which tasks benefit from AI and which don't.

The AI Fit Framework

Not every task benefits equally from AI. This framework gives you a fast way to score any task against its AI suitability — so you invest your time where the return is highest.

High fit

Repetitive & Structured

Data entry, formatting, scheduling, report generation, template-based communications

Good fit

Creative & Strategic

Brainstorming, first drafts, content ideation, presentation structuring, competitive research

Mixed fit

Relationship & Judgment

Negotiations, performance conversations, client management, conflict resolution

Low fit

Novel & Ambiguous

Strategic decisions with incomplete data, ethical dilemmas, culture shifts, crisis management

Where AI fits across business functions

Not theoretical — here's where professionals are getting measurable value from AI tools today. Start with one task and measure the difference before scaling.

  • Administrative — meeting summaries, data entry, scheduling, document formatting
  • Communication — drafting emails, editing for tone, translating, preparing talking points
  • Analysis — data summarisation, trend identification, competitive research, report generation
  • Creative — brainstorming campaigns, content ideation, presentation design, copywriting first drafts
  • Process — documenting procedures, creating checklists, onboarding materials, FAQ development
  • Research — market analysis, literature reviews, vendor comparison, regulatory summaries
Practical Exercise

Your AI Opportunity Map

Map your typical work week through the AI Fit Framework. Identify your top three opportunities and design a one-week test for the highest-potential one.

  1. 01List your 10 most time-consuming weekly tasks
  2. 02Score each: High / Good / Mixed / Low AI fit using the framework above
  3. 03Pick your top-scoring task. Use AI for it every day this week. Track time saved and quality difference
  4. 04After one week: keep it, modify your approach, or move to the next candidate
Takeaway

AI doesn't transform your job overnight. It transforms specific tasks. Finding the right tasks is the skill.

Module 05

Data, Privacy
& Security

The problem this solves

This is the module most people skip and most organisations wish their staff hadn't. The fastest way to create risk with AI is to use it carelessly with sensitive information — and most people don't realise what "sensitive" covers.

Where Does Your Data Go?

When you type something into a consumer AI tool, that data may be stored, used for training, and accessible to the provider's staff. Enterprise deployments offer data isolation. The difference isn't just features — it's data handling, retention, and access controls.

The Data Leakage Problem

Data leakage through AI isn't about hackers — it's about habits. Pasting client financials into a public tool for "a quick summary." Uploading confidential strategy docs for formatting. Each creates exposure that's invisible in the moment and potentially catastrophic later.

The Australian Context

Australia's regulatory landscape around AI is evolving. The government has signalled intent around AI governance, voluntary safety frameworks are in play, and Privacy Act reforms have AI implications. "We'll figure it out later" is a risk position, not a strategy.

Consumer vs Enterprise AI

Free tiers often have weaker data protections. Before using any AI tool for work, check: does your data get used for training? What's the retention policy? Where is it stored geographically? Who can access it? These aren't technical questions — they're risk management basics.

The "Never Paste" List

Regardless of which AI tool you're using, these categories should never go into a public or consumer-tier tool without explicit organisational approval. Check off each one as you review it.

Client data — names, financials, project details, communications
Personal identifiable information — staff or customer records
Financial information — accounts, forecasts, pricing models
Proprietary strategy — competitive plans, M&A, IP details
Legal documents — contracts, disputes, regulatory filings
HR records — performance reviews, salary data, medical info
Authentication credentials — passwords, API keys, tokens
Source code — proprietary algorithms, system architecture
Practical Exercise

The Data Sensitivity Triage

Five realistic workplace scenarios involving AI use with data of varying sensitivity. Classify each as Safe, Conditional, or Unacceptable. Click to reveal the assessment.

Scenario 1

Using AI to brainstorm subject lines for a marketing email about a public product launch.

Tap to assess →
✓ Safe. Public product, generic creative task, no sensitive data. This is exactly where consumer AI tools add value without risk.
Scenario 2

Pasting a client's quarterly financial summary into ChatGPT to generate talking points for a board presentation.

Tap to assess →
✗ Unacceptable. Client financial data in a consumer tool creates leakage risk. Use an enterprise tool with data isolation, or summarise manually and use AI only for structuring your own talking points.
Scenario 3

Using your company's enterprise AI platform to draft a performance review for a direct report.

Tap to assess →
⚠ Conditional. Enterprise platform helps, but check your org's policy — does it allow HR data in AI tools? Even enterprise tools may not have access controls covering performance data. Verify first.
Takeaway

Using AI responsibly isn't about restriction — it's about knowing which tool to use, how to use it, and what to keep out of it. This is professional hygiene, not bureaucracy.

Module 06

Building Your
AI Practice

The problem this solves

Most training creates a spike of enthusiasm that fades within a fortnight. This module exists to prevent that — turning what you've learned into a sustainable daily practice that compounds over time.

Build Your Personal Toolkit

Choose 2–3 AI tools that fit your role and commit to them. Depth beats breadth. A professional who's excellent with one AI tool will outperform someone who dabbles in ten. Learn the shortcuts, the quirks, the limitations — that's where real productivity gains live.

The 30-Day Habit Framework

Week 1: One AI task per day, same type. Week 2: Add a second task type, refine Prompt 1. Week 3: Three tasks, iterate instead of accepting first outputs. Week 4: Audit what's working, drop what isn't, lock in your practice.

Stay Current Without Drowning

Curate 2–3 reliable sources that filter the noise: one newsletter, one industry publication, one peer group. Fifteen minutes a week is enough to stay informed without becoming overwhelmed. The AI landscape moves fast, but you don't need to track every announcement.

Be the AI-Literate Colleague

Share what you learn with boundaries. Show one person one useful trick this week. Contribute to your org's AI maturity without becoming the unpaid support desk. Set expectations: "I can show you how I use it; I can't troubleshoot every tool."

Capstone Exercise

Your Personal AI Action Plan

This is your capstone deliverable for Track 1. Build a concrete, actionable plan that turns the past five modules into lasting change. No grand transformation — just small, specific commitments.

  1. 01Three tools — which AI tools will you commit to using? Name them specifically
  2. 02Three tasks — which work tasks will you apply AI to first? Be specific about the task and expected benefit
  3. 03One weekly habit — what's your 15-minute-a-week practice for staying current?
  4. 04One share — what will you show or teach one colleague within the next 7 days?
Takeaway

AI fluency isn't a destination — it's a practice. Small, consistent use beats occasional ambitious experiments every time.


What's next

Ready to go further?

Track 1 gives you the foundation. The next tracks build on it — moving from personal AI fluency to operational leadership and strategic decision-making.

Track 2: AI for Operations & Management  ·  Track 3: AI Strategy & Leadership

Talk to us about training