Outrage Filter
A browser and mobile layer that detects rage-bait, outrage framing, and emotional-escalation content, then dampens it with context, cooling delays, or quiet removal from your feed.
How It Works
The application runs on the user's device, applying directorate-developed models to detect and interrupt manipulative patterns in real time. All processing occurs locally — no personal data leaves the device. Anonymised aggregate telemetry (opt-in only) feeds back into the directorate's technique detection pipeline, improving the models for every user.
Who It's For
Individual users who want to reclaim control over their attention and digital habits. No technical expertise is required — the tools are designed for mainstream adoption.
Key Features
- Structured data feeds with documented schemas
- Privacy-first architecture — no personally identifiable data collected
- Regular updates aligned with the directorate's Technique Index releases
- Cross-platform compatibility and offline capability where applicable
Ethics & Provenance
Trained on the directorate's taxonomy of emotional-escalation patterns. The filter scores content for manufactured outrage signals — inflammatory framing, missing context, engagement-bait headlines — and softens or flags it before the amygdala fires.
Where the Money Goes
Revenue sustains the emotional-manipulation detection models and the public wellbeing tools.