# AI Bootcamp Follow-up ## From Awareness to Value Creation Leo Pharma — Global Clinical Development February 27, 2026 hey

# Today's Agenda ### Morning - Where are we now? - Value creation & measurement - Business cases & success stories ### Before Lunch - Prompting masterclass - AI tools for leadership - Live demos & next steps

# Since Last Time - AI moved from hype to infrastructure - Models improved — but **prompting skills** matter more than model choice - Copilot, Claude, Gemini — all accessible, all useful - The real gap: **potential vs. actual daily usage**

# The 2026 Shift > "We're no longer asking *should we use AI?* > We're asking *how do we get value from it?*" - From awareness → adoption - From experiments → embedded processes - From individual use → organizational capability

<!-- SECTION: Value Creation & Measurement --> # Value Creation & Measurement How do we create value? How do we know?

# Where Value Comes From ### Time Savings - Email drafting: 15 min → 2 min - Report summaries: hours → minutes - Meeting prep: automated briefs ### Quality Gains - Consistency across documents - Fewer revision cycles - Better first drafts

# What Successful AI Initiatives Have in Common - **Task-oriented:** Solve a specific, recurring pain point - **Measurable:** Before/after comparison is possible - **Low friction:** Integrated where people already work - **Champion-driven:** Someone owns it and shows others - **Domain-grounded:** AI + expert knowledge, not AI alone

# Do's and Don'ts ### ✅ Do - Start with tasks you repeat weekly - Combine AI output with your expertise - Share successes visibly (10-min case demos) - Measure before & after - Make AI part of existing processes ### ❌ Don't - Chase the newest model — skills matter more - Use AI for tasks requiring 100% accuracy without review - Keep it voluntary — embed it in workflows - Expect magic without context - Ignore data classification rules

# Discussion: GCD Today > Where are you seeing the biggest gap between > AI's potential and actual daily usage in your teams? - Which tasks consume the most time? - Where have you seen early wins? - What's blocking broader adoption?

<!-- SECTION: Prompting Masterclass --> # Prompting Masterclass Context + Prompting + Domain Knowledge

# The Real Success Formula > It's not about the model. > It's about **Context + Prompting + Domain Knowledge**. - **Context:** Give AI the background it needs - **Prompting:** Structure your request clearly - **Domain knowledge:** You are the expert — AI is the accelerator

# The 4-Block Prompt Framework | Block | Question | |-------|----------| | 🎭 **Role** | Who should the AI be? | | 🌍 **Context** | What background does it need? | | 📋 **Task** | What exactly should it do? | | 📐 **Format** | How should the output look? |

# Example: Weak → Strong Prompt ### ❌ Weak "Write a summary of the meeting." ### ✅ Strong "You are a senior clinical project manager. I had a 1-hour steering committee meeting about trial X. Here are my notes: [paste]. Write a structured summary with: decisions made, action items with owners, and open questions. Max 1 page, bullet points."

# Exercise: Improve This Prompt > "Help me write an email about the project delay." Rewrite it using the 4-block framework: **Role + Context + Task + Format** Try it in your AI tool now — compare before/after.

# AI Profiles & Agents - A **custom AI profile** = pre-loaded context for recurring tasks - Instead of re-explaining every time, build a reusable prompt - Examples for leaders: - Meeting summarizer (your format, your priorities) - Email drafter (your tone, your role) - Decision brief generator (your stakeholders, your metrics)

# Exercise: Build Your AI Profile > Write a system prompt that describes: > Who you are, what you do, what tone you prefer, > and what kind of tasks you'll ask AI to help with. Save it — this becomes your personal AI accelerator.

<!-- SECTION: AI for Leadership --> # AI Tools for Leadership Show it, don't tell it

# The Leadership Challenge > "You don't become good *because* of AI. > You become better *with* AI — if you already have the expertise." - AI is a junior colleague — fast but needs direction - Your domain knowledge is the foundation - Guide your teams to use AI where they're already strong

# Show It, Don't Tell It - **Use AI visibly** — in meetings, in emails, in prep - **Share your prompts** — normalize the workflow - **Invite 10-min case presentations** — let team members show their wins - **Embed in processes** — not as optional, but as standard - **Celebrate experiments** — even failed ones

# Three Layers of AI Adoption ### 🔝 Leadership / EVP Strong alignment — set the direction ### 🔄 Middle Management Critical gap — needs development ### 👤 Individual Users Good understanding — need permission & tools > The middle layer is where adoption succeeds or stalls. > Directors bridge top-down vision with bottom-up experimentation.

# Integration = Convenience = Adoption - The best AI tool is the one that's **already open** - Copilot in Outlook/Teams — low friction, high frequency - AI that lives in your workflow beats a separate tool - Think: Where do your teams spend their screen time?

<!-- SECTION: Demos --> # Live Demonstrations Claude Code · OpenClaw · LEO AI

# Demo: Claude Code - AI-powered coding and automation directly in the terminal - From idea to working solution in minutes - Real-world example: building tools, analyzing data, creating presentations *Live demo*

# Demo: OpenClaw - Multi-agent AI workflow orchestration - Specialized agents collaborating on complex tasks - From single prompts to automated pipelines *Live demo*

# Demo: LEO AI + PowerPoint - AI integrated into your daily tools - Creating presentations with AI assistance - Internal tooling perspective *Demo with Natasja*

<!-- SECTION: Next Steps --> # Next Steps From discussion to action

# Which 3 Use Cases Do We Tackle First? > Think about your team's biggest time sinks. > Which ones could AI meaningfully improve *this quarter*? - Criteria: High frequency + clear before/after + someone owns it - Let's identify the top 3 as a group

# Ownership & Metrics ### Who owns what? - **GCDLT:** Champion use cases, model behavior - **Adoption team:** Training, enablement, support - **IT:** Tools, access, compliance, data governance ### Simple metrics - Active users per department - Hours saved per use case - Tasks automated / month - Team satisfaction (quick pulse)

# Thank You ### Let's make AI work for GCD in 2026 Torben Madsen — Kvalitor Consulting [tm@kvalitor.dk](mailto:tm@kvalitor.dk)

<!-- REFERENCE: Syntax examples (delete this section before presenting) --> # Syntax Reference ### YouTube embed <div class="youtube-embed"><iframe data-src="https://www.youtube-nocookie.com/embed/dQw4w9WgXcQ" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen loading="lazy"></iframe></div>

# Syntax: Mermaid Diagram <div class="mermaid"> graph LR A[Identify Task] --> B[Build Prompt] B --> C[Run AI] C --> D{Good enough?} D -->|Yes| E[Use Output] D -->|No| B </div>

# Syntax: All Features - **Link:** [OpenAI](https://openai.com) - **Image:** `` - **QR code:** `<img src="https://api.qrserver.com/v1/create-qr-code/?size=300x300&data=https://example.com" alt="QR Code" style="max-height:300px;border-radius:8px">` - **Timer:** `<div class="timer-display" data-seconds="300" style="font-size:3em;color:var(--leo-accent);font-weight:bold">5:00</div>` — 5 min countdown - **Fragment:** add `<!-- .element: class="fragment" -->` after a bullet - **Speaker notes:** add `Note:` block at end of slide (press **S**) - **PDF export:** add `?print-pdf` to URL, then Ctrl+P - **Slide background:** `<!-- .slide: data-background="#003366" -->` - **Code highlight:** fenced code block with language