LLMs in Security: From Vulnerability Judgment to Hypothesis-Falsifying Agents

A security talk on replacing model confidence with executable evidence, stateful protocol fuzzing, and verification-driven agent workflows.

August 5, 2026
# AI Security# Vulnerability Research# Fuzzing# Security Agents
1 /
Loading presentation…

Presentation overview

This presentation argues that a plausible vulnerability explanation is not security evidence. Instead of treating an LLM as a final judge, it proposes using the model to generate hypotheses while execution, replay, tests, sanitizers, and observable effects determine whether a finding is real.

Topics covered

Core takeaway

The research target should be the verification system around the model, not the model’s confidence alone. A useful security agent narrows the search space, chooses the cheapest falsifying experiment, discards findings that cannot be reproduced, and operates under strict sandboxing, provenance, and least-privilege controls.