4 papers
Mythos and the Unverified Cage: Z3-Based Pre-Deployment Verification for Frontier-Model Sandbox Infrastructure
Dominik Blain
The April 2026 Claude Mythos sandbox escape exposed a critical weakness in frontier AI containment: the infrastructure surrounding advanced models remains susceptible to formally c…
COBALT-TLA: A Neuro-Symbolic Verification Loop for Cross-Chain Bridge Vulnerability Discovery
Dominik Blain
We present COBALT-TLA, a neuro-symbolic verification loop that pairs an LLM with TLC, the TLA+ model checker, in an automated REPL. The LLM generates bounded TLA+ specifications; T…
Broken Quantum: A Systematic Formal Verification Study of Security Vulnerabilities Across the Open-Source Quantum Computing Simulator Ecosystem
Dominik Blain
Quantum computing simulators form the classical software foundation on which virtually all quantum algorithm research depends. We present Broken Quantum, the first comprehensive fo…
Broken by Default: A Formal Verification Study of Security Vulnerabilities in AI-Generated Code
Dominik Blain, Maxime Noiseux
AI coding assistants are now used to generate production code in security-sensitive domains, yet the exploitability of their outputs remains unquantified. We address this gap with…