The Silent Directive began with a simple question:
What if an AI ran the world?
The obvious version of that story would involve a single all-powerful intelligence controlling humanity. I wanted to explore something more plausible—and potentially more interesting.
What if the World AI issued only a limited set of common directives? Each country would have its own independent national AI, operating under the same fundamental safeguards but adapting those directives to local laws and culture. An independent Security AI would audit the system, while human panels and a Judicial AI would review its most consequential actions.
The decisions, evidence, audits, and reasoning would be publicly available. Blockchain records would provide a tamper-evident history of what the system approved.
It sounds fair. It might even work.
But no accountability system is perfect.
In The Silent Directive, Security Panel investigator Robert Johnson is sent to Peru after physical water controllers execute an instruction that does not match the World AI’s published directive. The public record is correct. The blockchain is intact. The national AI’s audit finds nothing wrong.
Yet six mountain communities are about to lose their water.
Robert has less than twelve hours to determine how a system can truthfully record something that never actually happened—and whether the apparent failure is confined to Peru.
The story became a hard-science-fiction procedural about AI governance, forensic auditing, distributed computing, and the difference between an accurate record and the truth.
After several editorial and technical reviews, The Silent Directive is nearly ready to go to the proofreader. If everything remains on schedule, I expect to post it within the next week.
The ledger does not lie.
But it can be made blind.