dignitea

The Architecture of Foresight: What Makes a Prediction “Grounded”?

Koo Ping Shung

Dignitea, AI Consultancy & Training

As we settle into the first month of 2026, the tradition of the “Annual Prediction” is in full swing. Everyone from industry titans to casual observers is eager to offer their vision of what the next twelve months will hold for AI, the economy, and society. While I believe this is a healthy exercise for the mind, it has prompted me to ask a critical question: Whose predictions should we actually trust? In an era of information overload, how do we distinguish between a wild guess and a “grounded prediction”—one that is analytically assessed to be likely to happen?

The Anatomy of a Grounded Prediction

To evaluate the weight of a forecast, we must look at the mechanics of how it was constructed. Here is how I will identify a grounded prediction:

Extrapolation vs. Information Density

Most predictions are essentially an extrapolation of what the predictor has read or experienced thus far. However, there is a direct correlation between information density and accuracy. The more a person has read, researched, and personally experienced, the more “raw material” they have to work with. This increased volume of information acts as a filter, increasing the probability that their prediction is rooted in reality rather than just a headline-grabbing trend.

The Forecasting Parallel: Capturing the Nuance

Grounded predictions are remarkably similar to statistical forecasting. In forecasting, a single data point tells you very little, but a multitude of data points allows you to see the “signal” through the “noise.” A seasoned predictor uses their vast data set to capture the subtle nuances that shift a trajectory’s direction. They aren’t just looking at the straight line; they are looking at the small, overlapping forces that cause the line to curve.

The Experience Filter: Weighting the Variables

This is perhaps the most critical stage. A person who has been in a field for a long time develops an intuitive “weighting system.” They can determine if a captured nuance is a genuine disruptor or merely a temporary distraction. Their experience allows them to assess not just what might happen, but the magnitude of impact that specific variables will have on the final outcome.

The Missing Link: Pattern Recognition.

Beyond just having data, a grounded predictor recognizes historical patterns. They may be aware that technology often follows an “S-curve” of adoption, or that certain regulatory shifts have historically preceded market consolidations. This ability to map current events onto historical cycles is what prevents a prediction from being “over-fitted” to the present moment.

Conclusion:  Evaluate the Predictor, Not Just the Prediction

The “groundedness” of a prediction—and whether you should use it as a reference for your own business or life decisions—ultimately depends on the architecture of the person making it.

If you are a financial modeler, your advantage is not being able to crunch the numbers (an agent can do that); it is your ability to recognize if the model is fragile, maintainable, and reflects the true logic of the business. If you are a product manager, it is not being able to write a PRD (an agent can do that in 10 minutes); it is your ability to sniff-check the strategy for coherence, identify crucial gaps, and know which risks are worth taking.

The value we bring to society will be disproportionately determined by our ability to move to meta-skills. Evaluation competency now sits above execution competency.

Our advantage is our ability to think clearly, to decompose complex human problems into verifiable processes, and to use our internalised “taste” to act as a judge. We must do three things to thrive:

• Embrace the role of the Sniff-Checker. Focus on developing your taste, taste, and knowledge of best practices so you can quickly evaluate outputs.
• Learn to Build the Harness. The skills of the future involve managing and scaffolding AI agents—creating the handoffs, roles, memory systems, and verification procedures that allow them to work collectively on your terms.
• Decompose Your Own Work. Stop asking “Can AI do my job?” and start asking “Can my work be decomposed into verifiable sub-processes?” Map out your domain and identify the 80% you can offload, freeing yourself to focus on the 20% that requires true human judgment.

This is not a scenario of humans vs. machines. It is a world of the “team of one” managing a team of a hundred agents. The goal is to use our human judgment to bring agents into the space to extend our leverage, to act as the pilot of a high-output system rather than a manual laborer. We won’t be replaced; we will be upgraded.

More from Koo Ping Shung at koopingshung.substack.com