5#AIOfficerCertification | Dave Hajdu | dave@ai-officer.com
2 / 20
You Built the Mindset
1Mission 1: You learned to lead AI.
2Mission 2: You learn to fuel it.
3 / 20
The Ferrari Problem
1They bought the Ferrari.
2They filled it with swamp water.
4 / 20
Three Stories You Will Recognize
1The chatbot nobody used.
2The resume screener that missed everyone.
3The report nobody trusted.
5 / 20
Section Divider - How AI Actually Processes Data
1How AI Actually Processes Data
2Three layers. One that matters most.
6 / 20
The Three-Layer Sandwich
1Layer 1: The interface. (Where you work.)
2Layer 2: The LLM. (Where it thinks.)
3Layer 3: The data you feed it right now. (Where results happen.)
7 / 20
Three Layers of Data
1Training data: fixed. You cannot change it.
2Application memory: what the app remembers about you.
3What you feed it right now: the only layer you control.
8 / 20
Your Advantage Nobody Can Buy
1Anyone can buy the same tools.
2Nobody else has your data.
9 / 20
Exercise - The Two Halves of Your Data
1Exercise: Map Your Data Moat
2What data does your organization have that nobody else can replicate?
3Where does it live right now?
10 / 20
Section Divider - Safety and Governance
1Safety and Governance
2Know where your data goes before it goes there.
11 / 20
The Three Tiers
1Free: Your data trains the model.
2Paid: Your data stays yours.
3Enterprise: Your data has a contract.
12 / 20
Check Your Settings
1Exercise: Settings Audit
2Open your AI tool. Find the privacy settings. Answer three questions: Is my chat history being saved? Is my data training the model? Are my uploads stored?
13 / 20
Section Divider - The Right AI for the Right Job
1The Right AI for the Right Job
2Four criteria. Every time.
14 / 20
Four Selection Criteria
11. The data behind the model.
22. The interface features your team needs.
33. The safety controls required.
44. The apps it integrates with.
15 / 20
The Wrapper Warning
199% of AI apps are wrappers.
2Know what is under the hood.
16 / 20
Worked Example: Choosing a Tool for a Program
1Start with the outcome.
2End with the tool.
17 / 20
The BOLT Challenge
1Dana needs clean data.
2The BOLT launch is waiting.
18 / 20
Closing Reflection
1The leader who checks the numbers before they go to leadership is the leader who gets trusted with more numbers.
19 / 20
Your Commitment
1Start, stop, or continue.
2Name one behavior your manager would notice this week.