Steve Phillips, Cofounder and CEO, overseeing Zappi’s global business, product and growth strategies.
Most of us never reach for a map while driving anymore, nor are we printing directions before a family trip. We navigate with our phones, guided by the pinpoint accuracy of GPS. It gets us to vacation destinations, client meetings and hospital appointments without a second thought. But GPS wasn’t always this reliable.
The first satellites in the 1970s could only provide a fix within hundreds of meters. It was only when GPS opened up for wider civilian and commercial use, and when augmentation systems like Differential GPS were added, that accuracy improved to the level we now take for granted.
Unrealized Potential: What GPS Can Teach Us About AI
The increasing adoption and accuracy of modern digital GPS systems created new behaviors. We stopped treating physical maps as navigation tools. Instead, GPS became the invisible infrastructure that underpins how we move through the world. Entire industries—ride sharing apps, logistics, food delivery—emerged because of it.
Artificial intelligence sits in a similar place today. Large language models and generative AI have shown dazzling promise, but their weaknesses are well known. Hallucinations, overconfidence and lack of transparency pose real barriers to business use. Just as no pilot would once have flown an aircraft relying only on an early GPS fix, few leaders today would base a multimillion-dollar strategy on an unverified AI output.
And yet, the unrealized potential is enormous.
AI As A Navigation Tool For Businesses
If GPS transformed how we navigate the physical world, AI has the potential to transform how we navigate the internal world of decision making.
This is critical because decision velocity has exploded. Harvard Business Review notes that three-quarters of executives say their pace of decisions has increased tenfold in recent years. Leaders are inundated: more data, more dashboards, more KPIs, but also more noise, more silos, more competing versions of the truth. Teams across the organization are working with fragmented slices of data, seldom putting them together to paint the full picture of the consumer.
Technology so far has chipped away at the problem. Platforms have made data easier to access. Dashboards have democratized reporting across functions. But too often data remains fragmented, decision making processes remain linear, and the user experience is full of friction.
From Software To Systems
AI could mark the moment we move from a platform revolution to a systems revolution.
So far, technology has focused on building efficiency tools—faster dashboards, smarter platforms, more accessible data. But AI introduces something different: a connected, adaptive layer that can sense, interpret and respond across the whole system.
The ability to connect multiple datasets with a hive mind that can operate simple queries has the potential to guide leaders through increasing complexity.
The gap between this potential and the reality, however, is the training data of LLMs. AI tools can’t deliver on reliability when used in isolation because bias, inaccuracies and hallucinations could cloud their outputs.
When paired with connected consumer and business data, this is a flaw that can be resolved. When we infuse the hive mind with bespoke consumer and business data, accuracy and relevancy improve. For example, at Zappi, we’ve helped brands pair AI-augmented analytics with real consumer feedback to create connected systems that guide decision making across every stage of advertising development, from evaluating early ideas and refining storylines to validating final executions. This systemized approach enables teams to learn in real time, shorten decision cycles and build greater confidence in their creative choices, turning insight into a dynamic, continuous feedback loop.
The Huge Potential Of Connected Data When Paired With AI
AI paired with connected data could become the co-pilot for business strategy by accelerating decisions in real time, providing actionable insights at the point of need, and enabling synthetic testing and scenario modeling.
One practical way to think about connected insights is through a three-tiered approach. First, use generative AI models such as GPT, Llama or Claude as a base foundation. Second, add a middle layer of connected data, including real consumer data, synthetic data and trusted third-party sources, so the system is trained on relevant and accurate inputs. And finally, make that intelligence actually usable by simplifying access through an intuitive interface.
This structure can support continuous learning loops by enabling brands to adapt campaigns in the market, improve creative effectiveness and make faster decisions grounded in live data signals.
A Paradigm Shift In Consumer Data
This represents a shift from linear research projects to flowing, interconnected systems. Data from across the organization—insights, marketing, finance, sales—feeds into a single, unified environment.
Instead of sitting on the sidelines, insight teams become central nodes by connecting the dots between datasets and orchestrating how AI tools are applied. Decision making becomes less about static reports and more about continuous, living hubs of intelligence.
If GPS allowed us to move more confidently through the physical world, AI could allow businesses to move more confidently through a volatile, data-saturated world.
Because just like GPS, AI will not transform our reality overnight. It will take accuracy, trust and the willingness to reimagine systems, not just tools.
Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?