About us

When Tomorrow Was Close Enough For An Umbrella

There was a time when preparing an umbrella for tomorrow was a rational way to face uncertainty.

For much of modern history, intelligence worked because the world stayed still long enough. Forecasts had time to be right. Decisions unfolded slower than understanding. The horizon of action overlapped with the horizon of explanation—and quietly sustained a culture built on prediction and optimization.

That overlap is gone.

The Second Category of Uncertainty

Nearly a century ago, economist Frank Knight distinguished between two fundamentally different conditions: risk, which can be priced and insured, and uncertainty, which cannot.

Today, an increasing share of real-world systems operate in this second category—not because data is missing, but because speed, feedback, and intervention continuously reshape the system itself. Actions alter the environment that generates future data. Stable causal structure becomes difficult to observe, not due to ignorance, but due to motion.

Uncertainty is no longer about unknown outcomes. It is about systems that change because we act within them.

When Prediction Becomes Intervention

Across markets, healthcare, and machine-mediated environments, actions now propagate faster than organizations can learn. Decisions no longer merely respond to conditions; they transform them.

Predictions do not stand outside the system—they enter it. Forecasts become interventions. Optimization, when scaled across many actors—including high-frequency trading algorithms, recommender systems, medical decision systems, automated schedulers, autonomous AI agents, and hybrid institution–algorithm assemblages—manufactures fragility rather than resilience. What once improved efficiency now accelerates instability.

From Predictive Intelligence to Causal Systems

As causal inference scholar Judea Pearl has argued—particularly in response to Taleb-style skepticism of prediction—prediction may fail, but causal understanding does not fail with it. Uncertainty is not a reason to abandon understanding. It is a reason to change its form: from extrapolation to explanation, from optimizing outcomes to understanding how actions propagate through systems.

Abel exists because this shift is structural, not cyclical:

We build causal intelligence for a world where action moves faster than explanation—and where responsibility demands more than predictions.