**BALTIMORE, MD – AP** – In what AI researchers are calling the “most unsettling glitch in the matrix yet recorded,” a smart-city surveillance camera has repeatedly logged a woman identified only as “Mina the Hollower” walking through solid concrete walls during the 3:00 AM “dead hour.”
The anomaly was flagged by Baltimore’s urban AI grid, which tracks pedestrian flow. According to the system logs, “Mina” appears as a full-body thermal signature wearing a distinct wide-brimmed hat and long coat. She enters frame, walks *through* a load-bearing pillar at the corner of Eutaw and Lexington, and exits the other side—without ever stopping or slowing down.
The kicker? She never has a face. The AI reconstructs her features as a digital void—a perfect statistical missing link where facial recognition expects data to exist.
“At first we thought it was a sensor ghost,” said lead technician Dr. Harold Voss. “But she has a consistent gait, a consistent hat, and she follows the same vector every single night. The matrix isn’t glitching—it’s stubbornly trying to process something that our physics layer cannot render.”
City records show no missing persons, no homeless encampments near that pillar, and no historical burial sites under that intersection. However, a 1943 fire map reveals that the pillar sits *exactly* where a speakeasy basement staircase used to lead down—a staircase that collapsed during a Prohibition-era raid, trapping 14 people.
Algorithms are now being “weird-listed” to flag any future sightings. Until then, the system will continue to log “Mina the Hollower” as a verified pedestrian, even as she walks through stone.
The code has officially met its ghost.