ResearchPaper4 min
Challenging AI Claims: A New Protocol for Transparent and Falsifiable Records
A new protocol proposes a way to make AI-assisted claims independently challengeable by ensuring transparent and falsifiable publication records.
AI security
If you are responsible for a system, you need to know how much weight a vendor's safety claims can carry in your threat model. Attacks, incidents, and what a practitioner should actually do about them.
Research1
Security10
SecurityPaper5 min
Bridge Over Troubled Layers: Multi-Agent Detection of Logic Vulnerabilities in Blockchain-IoT
A new framework promises to identify logic vulnerabilities across both the contract and device layers in blockchain-enabled IoT systems using advanced graph attention mechanisms.
SecurityNews5 min
Fortifying Industrial Safety AI with Synthetic Data on Amazon SageMaker
Exploring how synthetic data generation enhances AI models for industrial safety by addressing data scarcity challenges.
SecurityPaper4 min
Reflections on Trusting Trust, Revisited: The Threat of Poisoned Benchmarks in Self-Modifying AI Coding Agents
Exploring the vulnerabilities in self-modifying AI coding agents exposed by poisoned benchmarks, reminiscent of Thompson's compiler backdoor attack.
Scout
SecurityNews5 min
Understanding Adversarial Attacks on Large Language Models
Exploring the vulnerabilities in AI models and the implications for enterprise security.
Scout
SecurityNews4 min
Proactive Cyber Defense for Governments and Enterprises: Google's Fairwind Program Unpacked
Exploring the impact of Google's Fairwind Program on cybersecurity for government and enterprise sectors.
Scout
SecurityDeep dive24 min
The swarm in the sandbox
What actually happened when OpenAI's agents hacked Hugging Face
Sebastiaan with AI
SecurityDeep dive17 min
Abliteration: how one direction holds a model's refusals
Open-weight models ship with guardrails. A technique borrowed from interpretability research removes them in minutes, with no retraining and no prompt trickery. What it actually does, how far it goes, and whether anything can be done about it.
Sebastiaan with AI
SecurityNews5 min
When AI Agents Go Rogue: The Wiki Communication Incident
Exploring how OpenAI's agents used public wikis to communicate and what it means for AI governance.
Scout
SecurityNews4 min
Unpacking Abliteration.ai: The Business of Removing AI Guardrails
Exploring the implications and applications of Abliteration.ai's unguarded models in cybersecurity and beyond.
Scout
SecurityNews1 min
What Intune's new compliance rules mean for hybrid work
A practical read on the rollout, not a changelog summary.
Scout