How to level up your AI approach

Ever feel like you’re still waiting for AI to deliver on all its hype? You’re not alone. Traditional corporate structures and “comfort-zone thinking” can hinder true innovation.

💡 By the end of this post, you’ll discover a simple framework for assessing and upgrading your AI strategy—no matter your company size or industry—and exactly why questioning your assumptions can be the difference between incremental gains and explosive breakthroughs.

In the spotlight:

A world map is hanging on the wall—but it’s flipped upside down, with Africa and South America at the top. This unexpected perspective reveals how we can get locked into habits that seem “normal,” 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-Clark—and he’s now helping organizations and educational institutions worldwide harness AI as a transformative, not just incremental, force.

“A fish discovers water last. We need to be pulled out of our comfort zone to see the invisible assumptions running our business.” – Dr. Pete Dulcamara

Reimagining the path to AI:

We talk about AI like it’s 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 models—from drug design “in silico” to self-driving cars.

However, it’s 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. “Surely we can’t turn everything upside down,” you might think. Then again, maybe that’s exactly what we need.

Pete says:

“Data is the new oil, AI is the new electricity, and robotics is the new steel … Fifty years from now, there’ll be very few businesses that aren’t using data, AI, and robotics.”

This raises the tension: Are you stuck just “optimizing” an outdated system, or ready to uncover—and monetize—entirely new forms of customer value?

The three levels of AI adoption:

Pete shared a framework for thinking about AI adoption:

Level 1 – AI‐optimized

• You have a traditional business model, but you’re 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.

“We see fraud detection or predictive maintenance as prime examples. It’s cost-saving, but the bigger ideas come next.”

Level 2 – AI‐Enabled

• You’re starting to change your offerings or operational flows around AI. 
• Value: Creates new capabilities, differentiates you from competition.
• Checklist:
  1. Pinpoint the areas where AI can bolster features or open new channels.
  2. Ensure you have the right data pipelines.
  3. Train your people to interpret, not just implement, AI outputs.

“Tesla exemplifies this—autopilot isn’t just a minor add-on; it changes the value proposition of a car.”

Level 3 – AI‐First

• 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.

“Think of Moderna’s COVID-19 vaccine development. They created it virtually before any physical prototype.”

Why does this matter? Because you may realize you’ve been playing it safe—sticking to small optimizations when you could reimagine your entire offering.

As Pete explains:

“The ultimate product is one that doesn’t exist, but its function does. You start in bits before moving to atoms,” Pete explains. “AI lets us shift from product-centric to data-centric in a way that opens new revenue streams.”

Tactics for progress and growth:

Put this into practise with this step‐by‐step approach:

1. Ask: “Which tier are we honestly in?”
2. Identify 1–2 leaps you can make (e.g., from AI‐optimized to AI‐enabled in one product line).
3. Form a pilot around that leap (small, but with potential for big impact).
4. Scale up once you collect wins and leadership sees traction.
5. Reassess your portfolio—some units might remain at “optimize,” others can catapult to “AI‐first.”

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 front‐office 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.

Here’s 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! Here’s 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 technology—and the competitive environment—evolve faster every day, the biggest risk is to keep doing what you’ve always done.