- AI Unhyped
- Posts
- š How to level up your AI approach
š How to level up your AI approach
A Fortune 500 Chief Scientist reveals the three levels of AI adoption.

Hi, and happy Tuesday.
Ever feel like youāre still waiting for AI to deliver on all its hype? Youāre not alone. Traditional corporate structures and ācomfort-zone thinkingā can hinder true innovation.
š” By the end of this post, youāll discover a simple framework for assessing and upgrading your AI strategyāno matter your company size or industryāand exactly why questioning your assumptions can be the difference between incremental gains and explosive breakthroughs.
PreScouter Executive Roundtables
Join our complimentary membership program offering exclusive, live roundtables focused directly on each newsletter topic. Upcoming topics and Roundtable dates include:
How to Build AI-Driven Data Intelligence That Works - Tue 29th April
AI + How to Triangulate Suppliers Amid Tariffs - Tue 6th May
How to Get to the 95% of Knowledge AI Misses (via Experts) - Tue 13th May
How to Address AIās Biggest Blind Spot: Local Context - Thu 22nd May
(All sessions are held at 10am PT / 1pm ET / 7pm CET)
IN THE SPOTLIGHT

A world map is hanging on the wallābut itās flipped upside down, with Africa and South America at the top. This unexpected perspective reveals how we can get locked into habits that seem ānormal,ā yet limit what we believe is possible.
In a recent conversation with Dr. Pete Dulcamara, former Chief Scientist at Kimberly Clark, we explored how this shift in perspective applies directly to AI in business. He spent years leading global R&D at Dow and guiding massive projects at Kimberly-Clarkāand heās now helping organizations and educational institutions worldwide harness AI as a transformative, not just incremental, force.
āA fish discovers water last. We need to be pulled out of our comfort zone to see the invisible assumptions running our business.ā
Reimagining the Path to AI
We talk about AI like itās a fancy add-on, but remember how electricity revolutionized every corner of industry? AI is following the same path. Instead of merely automating what humans do, AI can spark entirely new business modelsāfrom drug design āin silicoā to self-driving cars.
However, itās not as simple as downloading a tool or hiring a data scientist. Many manufacturing or consumer-goods companies run on decades-old machinery and siloed structures that block data flow. āSurely we canāt turn everything upside down,ā you might think. Then again, maybe thatās exactly what we need.
Pete says:
āData is the new oil, AI is the new electricity, and robotics is the new steel ā¦ Fifty years from now, thereāll be very few businesses that arenāt using data, AI, and robotics.ā
This raises the tension: Are you stuck just āoptimizingā an outdated system, or ready to uncoverāand monetizeāentirely new forms of customer value?
The Three Levels of AI Adoption
Pete shared a framework for thinking about AI adoption:
Level 1 ā AIāOptimized
You have a traditional business model, but youāre injecting AI for efficiency or improved processes.
Value: Cuts cost, improves speed/quality.
Pitfall to Avoid: Getting stuck optimizing only, missing the chance to create new revenue streams.
āWe see fraud detection or predictive maintenance as prime examples. Itās cost-saving, but the bigger ideas come next.ā
Level 2 ā AIāEnabled
Youāre starting to change your offerings or operational flows around AI.
Value: Creates new capabilities, differentiates you from competition.
Checklist:
Pinpoint the areas where AI can bolster features or open new channels.
Ensure you have the right data pipelines.
Train your people to interpret, not just implement, AI outputs.
āTesla exemplifies thisāautopilot isnāt just a minor add-on; it changes the value proposition of a car.ā
Level 3 ā AIāFirst
Your entire business model or major product lines are born in AI.
Value: Can yield exponential growth or entire new markets.
Organizational Impact: Usually requires reorganizing teams, forging new partnerships, or adopting new business metrics.
āThink of Modernaās COVID-19 vaccine development. They created it virtually before any physical prototype.ā
Why does this matter? Because you may realize youāve been playing it safeāsticking to small optimizations when you could reimagine your entire offering.
As Pete explains:
āThe ultimate product is one that doesnāt exist, but its function does. You start in bits before moving to atoms,ā Pete explains. āAI lets us shift from product-centric to data-centric in a way that opens new revenue streams.ā
Tactics for Progress and Growth
Put this into practise with this stepābyāstep approach:
Ask: āWhich tier are we honestly in?ā
Identify 1ā2 leaps you can make (e.g., from AIāoptimized to AIāenabled in one product line).
Form a Pilot around that leap (small, but with potential for big impact).
Scale Up once you collect wins and leadership sees traction.
Reassess your portfolioāsome units might remain at āoptimize,ā others can catapult to āAIāfirst.ā
These are some common challenges you may face
Data Silos: The biggest stumbling block for going from optimization to new AI products.
Talent Gaps: Culture shock in R&D or frontāoffice teams who fear AI will replace them.
ROI Pressure: Large investments, uncertain payoff. Overcome via smaller pilots that show measurable gains.

Ready to level up your AI?
This prompt will help give you apply this concept to your own work and organization.
Consider these three AI Levels: (1) AIāOptimized: You have a traditional business model, but youāre injecting AI for efficiency or improved processes. (2) AIāEnabled: Youāre starting to change your offerings or operational flows around AI. (3) AIāFirst: Your entire business model or major product lines are born in AI. For <product, business process or department> at <company>, please brainstorm some opportunities at each of these AI Levels.
Hereās highlights of what I got when I provided packaging at a consumer packaged goods company. I love it when I get compliments from the AI. Maybe you will too?
Great prompt! Hereās a breakdown of opportunities at each AI Level:
AI-Optimized
Demand Forecasting: Use AI to predict how much packaging material will be needed, reducing waste and stockouts.
Quality Inspection via Computer Vision: Automate defect detection (e.g., print errors, seal integrity) on packaging lines using AI-powered cameras.
Optimized Packaging Design for Cost and Sustainability: Use generative AI to explore lower-cost or more eco-friendly designs based on constraints like durability or shelf-life.
AI-Enabled
Dynamic Label Customization: Enable packaging lines to print labels with localized promotions or personalized content using AI-driven segmentation.
Smart Packaging Selection Engine: AI tool that selects the optimal packaging type based on customer location, weather, product sensitivity, etc.
Consumer Sentiment-Driven Design Updates: Use NLP on reviews and social media to influence packaging refresh cycles or messaging.
AI-First
Interactive Smart Packaging: QR/NFC-enabled packages that use AI to deliver hyper-personalized experiences, instructions, or loyalty offers based on who scans it.
Autonomous Design Co-Creation: Launch a consumer-facing AI that co-designs packaging (colors, fonts, textures) with users in real-time, turning them into co-creators.
Circular Packaging Intelligence: A closed-loop system where packaging embedded with smart tags reports usage, damage, or location to optimize reuse/recycling.
Because technologyāand the competitive environmentāevolve faster every day, the biggest risk is to keep doing what youāve always done.
If you feel inspired, share this newsletter with a colleague to start a conversation with them on this topic.
Best,
Dino
From Our Interview Series with Pete Dulcamara
šÆ What youāll learn:
|
šÆ What youāll learn:
|
šÆ What youāll learn:
|
My Interview Notes
I came up with 7 proven frameworks straight from my chat with Pete Dulcamara for creating your AI strategy:
|