QIF-T0065

Algorithmic psychosis induction (recommendation weaponization)

A recommendation algorithm profiles user psychological vulnerabilities via behavioral data (watch time, engagement patterns, emotional triggers), then systematically serves content designed to destabilize cognitive function. The attack does not require a BCI — it operates through normal sensory channels (screen → eyes → visual cortex → limbic system → prefrontal cortex). The algorithm acts as both the reconnaissance tool (profiling) and the delivery mechanism (curated feed). Documented real-world outcomes include induced psychosis, radicalization, eating disorders, and suicidal ideation in vulnerable populations. With a BCI, this attack becomes catastrophically more effective: neural state data replaces behavioral proxies, stimulation bypasses conscious filtering, and real-time feedback loops enable millisecond adaptation. This technique represents the bridge between classical social engineering and neural-direct cognitive exploitation.