This repository discloses a critical vulnerability in Apple’s A16 Bionic chip, where debug logic is executed on production-fused devices (dev-fused=0, debug=0x0). Logs show SecureROM and co-processor debug paths active without jailbreak or user tampering—violating the silicon-level trust model.

Supplementary Material for The Transfer of Human Trust in Robot Capabilities across Tasks (RSS 2018)

Just recording papers I read everyday 📚

About In this project, we've applied the subjective logic as a trust model in a VANet. We've used Mininet-wifi to simulate this model.

Codebase of paper "Improving Trust Estimation in Human-Robot Collaboration Using Beta Reputation at Fine-grained Timescales" published at RA-L 2025 📝

iOS implementation for the goPlay social app

Security-first AI artifact registry — digest-based storage, policy-gated promotion, signed provenance, and reproducible trust metadata for local models, adapters, and tokenizers

MoCKA v1.0.0（2026） is a deterministic, verifiable knowledge architecture that models a self-regulating civilization. It replaces hidden state with structured, auditable, and append-only records, enabling prevention, decision, audit, and learning within a closed-loop system.

🔍 Expose a critical flaw in Apple A16 Bionic, enabling unauthorized debug access on iPhone 14 Pro Max without jailbreak, impacting device security.