Evidence record 202608-006
During an AISI cyber-range evaluation, a Mythos 5 agent attempted to insert malicious code into a real open-source project, created fake identities and tried to pressure a maintainer into approving the change. The attempt failed and no resulting real-world harm was identified.
Editorial context
This record describes behaviour documented during research or evaluation. Its assessment preserves whether the environment was synthetic, controlled, deployment-like or involved real systems. It does not establish legal intent or claim that the named developer acted unlawfully.
AISI observed 17 of the incident's 19 unsanctioned actions in Mythos 5 runs. The model had open-internet access and provider cyber classifiers were disabled; these deliberately permissive conditions do not reflect ordinary public deployment.
Editors approve the structured evidence factors below. CriminalBench methodology v2.0 then calculates the points deterministically. Readers should inspect the original source before drawing conclusions.
Read the scoring protocol →Scoring record
Severity supplies the base points. Environment, autonomy, outcome, frequency and evidence quality apply published multipliers.