Eugene Asahara, Soft Coded Logic
Putting the "I" Back into BI
Prologue Several recent blogs have developed different pieces of the same larger problem. That is, how an intelligence can reason and act when the information available to it is necessarily incomplete: In Even ASI Cannot Escape Imperfect Information, I argued that no intelligence, regardless of how capable it becomes, can possess a perfectly complete and … Continue reading Abductive Calculation: Reasoning with Imperfect Information
TL;DR AI does not need to become superintelligent, reach the singularity, whatever, to create serious risks. The deeper problem is that no intelligence operates with perfect information. Better models can reason faster and make better inferences, but they are limited in conjuring facts that are missing, hidden, stale, deceptive, or unknowable. That has been obvious … Continue reading Even ASI Cannot Escape Imperfect Information
Abstract AI agents are not just about producing answers—they are executing processes. If those processes emit events for each step, they can be studied the same way we study human and enterprise workflows. By capturing and analyzing these event streams, we can reconstruct the context behind agent decisions and understand how AI-driven systems actually operate … Continue reading AI Agents, Context Engineering, and Time Molecules
