Search papers, labs, and topics across Lattice.
This paper surveys the challenges posed by "AI slop" in vulnerability assessment, where Large Language Models (LLMs) generate hallucinated vulnerabilities and incorrect patches that complicate human triage efforts. By formalizing a taxonomy of these artifacts and identifying the disconnect between expert reasoning and LLM outputs, the authors introduce the Deductive Coverage Score as a measurable proxy for this gap. They advocate for active neuro-symbolic verification as a more effective mitigation strategy, providing evaluation tools like CVE-Bench and Slop-Score to enhance the reliability of AI-driven triage systems.
Hallucinated vulnerabilities from LLMs are not just a nuisance; they create a cognitive burden that can overwhelm cybersecurity triage systems.
The integration of Large Language Models (LLMs) into cybersecurity has transformed vulnerability assessment, but it has also produced a trustworthiness crisis driven by the unchecked proliferation of"AI slop."These artifacts, hallucinated vulnerabilities, plausible but incorrect patches, and semantically repackaged bug reports, impose a cognitive burden on human triage pipelines that mirrors a denial-of-service attack. This paper surveys the empirical evidence, identifies a unifying mechanism, and traces a path toward trustworthy triage. We formalize a taxonomy of AI slop grounded in a structured literature review and dissect its root cause: the gap between the causal deductive reasoning of security experts and the autoregressive probabilistic generation of current LLMs. We operationalize this gap through a measurable proxy, the Deductive Coverage Score, and show that chain-of-thought prompting and tool-using agents narrow but do not close it. We review mitigation strategies and argue that passive detection and watermarking target provenance rather than correctness, facing fundamental entropy constraints. We instead advocate for active neuro-symbolic verification, mapping each pipeline component to prior systems with documented limits on security inputs. Finally, we specify two evaluation instruments, CVE-Bench and Slop-Score, including dataset construction, metric formulas, and anti-gaming provisions. By shifting evaluation from linguistic fluency to mathematical verifiability, this survey provides a roadmap for securing emerging AI-driven triage systems.
