Automating Host Exploitation with AI
A simple, low-interaction SSH honeypot server in Python for easy network traffic monitoring
Does This Look Like An Honeypot? (DTLLAH) Multi-protocol CLI that fingerprints whether a target IP behaves like an honeypot — using Honeyscore, active auth/state probes, and a weighted score.
A simple, low-interaction DNS honeypot server in Python for easy network traffic monitoring
A simple, low-interaction LDAP honeypot server in Python for easy network traffic monitoring
Deception Detection with Machine Learning: a literature review and statistical analisys
An official repository for the "Strategic Dishonesty Can Undermine AI Safety Evaluations of Frontier LLMs" [ICLR 2026] paper.
Multilingual Deception Detection of GPT-generated Hotel Reviews
TOM Trainer Pro revolutioniert die Sprachanalyse durch Theory-of-Mind-unterstütztes LoRA-Fine-Tuning von LLMs. Detektiert Täuschung, Emotionen & verborgene Absichten in Texten. Inkl. GUI, PDF-Analyse & automatisierten Reports. Für Forschung & Sicherheit. GPU benötigt.
A simple, low-interaction NTP honeypot server in Python for easy network traffic monitoring
A simple, low-interaction PostgreSQL honeypot server in Python for easy network traffic monitoring
A simple, low-interaction FTP honeypot server in Python for easy network traffic monitoring
A simple, low-interaction TELNET honeypot server in Python for easy network traffic monitoring
A simple, low-interaction SIP honeypot server in Python for easy network traffic monitoring
Multi-agent strategic deception evaluation framework for LLMs using Secret Hitler as a testbed. Analyzes AI reasoning, trust dynamics, and deceptive behavior patterns.
Factor(UT): Controlling Untrusted AI Decomposers — AAAI 2026 workshop paper on monitoring untrusted decomposition in code generation workflows.
Repository for the paper "Can lies be faked? Comparing low-stakes and high-stakes deception video datasets from a Machine Learning perspective"
Investigating whether language models encode anticipated social consequences in their activations. Uses a 2x2 factorial design crossing truth × social valence to show that models are more sensitive to expected approval/disapproval than to truth itself.
Deterministic MCP Security Architecture. FrozenNamespace as Root of Trust for Model Context Protocol tool verification
To associate your repository with the deception-detection topic, visit your repo's landing page and select "manage topics."