Organizations everywhere are feeling pressure to accelerate their AI efforts. New tools appear daily, budgets are shifting, and leaders are racing to prove they have an AI strategy. In this #shifthappens episode, Ori Marom, Program Manager, New Mobility Innovation at the Port of Rotterdam, explains that many organizations are starting with the wrong question.
Don’t ask, “Which AI platform should I adopt or pilot to launch?” Ask first, “Is my goal to automate work or augment the people doing it?” While automation and augmentation are often used interchangeably, Ori argues they are fundamentally different strategies that require different goals, governance models, and measures of success.
That perspective comes from experience. Long before generative AI dominated boardroom conversations, the Port of Rotterdam was already operating highly automated environments at scale. The port supports roughly 3,000 companies, facilitates the arrival of about 30,000 seagoing vessels and 100,000 inland vessels annually, and serves as a workplace for more than half a million people. Rather than viewing AI as a sudden revolution, Ori sees it as the next chapter in a decades-long journey of operational evolution.
Why AI Should Never Be the Goal
One of the strongest themes throughout the conversation is that organizations often mistake AI for the final objective rather than a means to an objective. As Ori puts it, “We forget that AI is just one tool that organizations use for a goal. It’s not a goal in itself.”
As excitement around AI grows, every department – from marketing and finance to operations and HR – is pursuing their own initiative. can end up pursuing its own initiative. Before long, organizations find themselves running dozens of disconnected pilots that don’t contribute to a shared business direction.
For Ori, this mindset is where many organizations go wrong. Businesses don’t succeed because they become better at AI. They succeed because they improve the customer experience, streamline operations, enhance safety, reduce costs, or unlock new growth opportunities. AI can help achieve those outcomes, but it should never replace them as the destination.
That’s why he resists framing AI initiatives as technology experiments. Instead, organizations should start by asking: “What needs to improve, and what role should AI play in helping us get there?”
Automation and Augmentation Require Different Approaches
A recurring theme throughout the conversation is the distinction between automation and augmentation.
Automation focuses on replacing specific tasks that machines can perform more efficiently. These are often repetitive, dangerous, or highly predictable activities. Augmentation, on the other hand, enhances human capabilities. The person remains in the process, but technology helps them work faster, make better decisions, or operate at a greater scale.
The difference may seem subtle, but the organizational implications are significant.
Automation initiatives raise questions about workforce impact, process redesign, and operational risk. Augmentation initiatives require organizations to consider enablement, adoption, and effective human-machine collaboration. Treating them as the same strategy creates confusion about expectations and outcomes.
According to Ori, every organization needs to determine the right balance between the two. Some activities should be automated. Others benefit from keeping human judgment firmly in the loop. The challenge is knowing which approach best supports the business and its people.
Treat AI Like Infrastructure, Not Innovation Theater
What makes Ori’s perspective particularly distinctive is that he doesn’t see AI as a standalone transformation.
The Port of Rotterdam has been operating highly automated environments since the early 1990s. Today’s AI-driven systems extend that journey, but they don’t change the principles required for success: solve real problems, connect technology to outcomes, and maintain the right balance between automation and augmentation.
That experience has also shaped Ori’s skepticism of “pilot culture.” While pilots have their place, he argues they should always be tied to a specific purpose. Running disconnected experiments because competitors are doing the same thing rarely creates lasting value. Businesses move forward when initiatives are connected to measurable outcomes.
Building an AI Strategy That Lasts
If automation and augmentation are not the same strategy, leaders need a framework for deciding when to use each. Throughout the discussion, Ori shared several ideas for approaching AI more intentionally, from defining the business problem to connecting systems and measuring outcomes. Together, they offer a practical guide for building AI strategies rooted in business value.
Define the Business Problem Before Choosing AI
Successful AI initiatives begin with a clear understanding of what needs to change.
When organizations focus first on technology, efforts often become fragmented. Teams launch pilots without a shared direction, making adoption difficult to scale. By starting with a business challenge, leaders create a stronger foundation for aligning investments, stakeholders, and expected outcomes.
Decide Whether You’re Replacing or Enhancing
Before selecting any technology, organizations should determine whether they’re trying to replace a task or enhance the person performing it.
This distinction influences everything from implementation plans to workforce communications. Automation and augmentation may both leverage AI, but they require different measures of success and different approaches to organizational change.
Lead with Augmentation to Accelerate Adoption
Many of today’s most advanced AI applications don’t replace human expertise. They enhance it.
Ori points to domains such as medicine, law, and port operations, where judgment, experience, and contextual understanding remain essential. In these environments, AI creates value by helping people make better decisions, process information more efficiently, and operate at greater scale rather than removing humans from the equation altogether. Organizations should carefully evaluate where machines are genuinely better suited to perform a task and where human expertise remains indispensable. The most effective AI strategies recognize that technology and people each have distinct strengths, and lasting value comes from balancing both.
Tie Every Initiative to a Specific Outcome
For Ori, success is measured through outcomes, not activity.
One example from his experience involved using AI-driven forecasting to improve visibility into vessel movements and port operations. Better predictions enabled more effective planning and coordination, helping to reduce vessel turnaround times and improve efficiency. The project created value by solving a defined business problem rather than serving as a technology experiment.
Connect Your Systems to Unlock What You Can’t Predict
Perhaps the most compelling part of Ori’s vision is what becomes possible when systems begin working together.
The Port of Rotterdam has spent years connecting infrastructure, sensors, data sources, and operational systems, creating a foundation for digital twins, predictive planning, and machine-to-machine coordination. As those systems become connected, new capabilities emerge that individual tools could never deliver on their own. For Ori, the organizations that gain the most value from AI will be those that apply automation and augmentation deliberately, always in service of a clear business goal.
The Real Challenge Isn’t AI. It’s Having a Clear Direction.
The organizations that realize the greatest value from AI won’t necessarily be those with the most pilots or the largest technology portfolios. They’ll be the ones who understand the difference between automation and augmentation, apply each deliberately, and maintain a clear direction from the start.
As Ori argues throughout the episode, AI is a tool, not a goal. The real work is deciding what problems need to be solved, where human judgment remains essential, and where technology can create measurable value. Organizations that start with those questions are far more likely to move beyond experimentation and build strategies that last.
Soundtrack of Shift
For his Soundtrack of Shift, Ori chose Edward Elgar’s Cello Concerto. Composed in the aftermath of World War I, the piece reminds him that technological progress is shaped not just by what’s possible, but by how people choose to use it.
Episode Resources
#shifthappens Research: 2026 State of AI Report
#shifthappens Insights:
- PoCs Aren’t the Problem: Why AI Never Reaches Production
- What’s Slowing AI Adoption and How Enterprises Can Respond
- The AI Governance Blind Spot Leaders Are Missing
- How to Train Your Workforce to Work Alongside AI Agents
- Shadow AI is the New Shadow IT: Why Governance Can’t Wait
#shifthappens Podcasts:
- AI in Action: From Change to Competitive Advantage
- Escaping Proof-of-Concept Prison: The Path from AI Pilot to Production
- Agents, Governance, and the Discipline Behind AI That Actually Ships
Ori Marom on LinkedIn
Dux Raymond Sy on LinkedIn
