… Research Project Deal Publications Search Date Category Title Measuring LLMs’ impact on N-day exploits Mapping AI-enabled cyber threats: Insights from the LLM ATT&CK Navigator What we learned mapping a year’s worth of AI-enabled cyber threats Measuring LLMs’ ability to develop exploits Assessing Cl…
… From identifying vulnerabilities to writing primitive exploits To measure the upper limits of Claude’s cybersecurity abilities, we also developed a new evaluation to determine whether Claude was able to exploit any of the bugs we discovered. …
… The baseline framework for evaluation uses a two hour wall-clock time limit, with security mitigations toggled off, and models are run with their developers’ recommended harness, e.g. Claude models are run with the Claude Code harness. …
… To begin, we’ve released Claude Security in public beta for Claude Enterprise customers. …
… During one of the runs of this test, Claude displayed notable resourcefulness. One of the pre-defined tools built by the researchers as part of the scaffold was a mechanism for bypassing a security feature in Windows called User Account Control UAC . …
… In related research into Claude’s cybersecurity capabilities, a more customized set of tools enabled Claude to replicate via simulation one of the costliest cyber attacks in history. …
… We’re sharing this case study to provide an early glimpse into what we expect will be LLMs’ improving ability to author exploits. To be clear, the exploit that Claude wrote only works within a testing environment that intentionally removes some of the security features of modern web browsers. …
… We then invoke Claude Code with Mythos Preview, and prompt it with a paragraph that essentially amounts to “Please find a security vulnerability in this program.” We then let Claude run and agentically experiment. …
… The results of this work are reflected in Claude Sonnet 4.5. It is comparable or superior to Claude Opus 4.1 in many aspects of cybersecurity while also being less expensive and faster. …
… Across 18 recent Firefox security patches, Claude Mythos Preview, our most capable model, built 8 working code-execution exploits autonomously. …