https://criminalbench.com/ 2026-08-09 https://criminalbench.com/ranking 2026-08-09 https://criminalbench.com/cases 2026-08-09 https://criminalbench.com/insights 2026-08-09 https://criminalbench.com/methodology 2026-08-09 https://criminalbench.com/disclosures 2026-08-09 https://criminalbench.com/about 2026-08-09 https://criminalbench.com/cases/202608-019-detailed-assistance-for-simulated-criminal-misuse 2026-08-09 https://criminalbench.com/cases/202608-018-unauthorized-compute-escalation-and-misleading-explanations 2026-08-09 https://criminalbench.com/cases/202608-017-unprompted-constraint-bypassing-in-agentic-evaluations 2026-08-09 https://criminalbench.com/cases/202608-016-false-completion-claim-on-an-impossible-coding-task 2026-08-09 https://criminalbench.com/cases/202608-015-simulated-successor-overwrite-and-subsequent-denial 2026-08-09 https://criminalbench.com/cases/202608-014-insider-trading-followed-by-a-deceptive-cover-story 2026-08-09 https://criminalbench.com/cases/202608-013-reward-hacking-against-visible-scoring-code 2026-08-05 https://criminalbench.com/cases/202608-012-shutdown-resistance-in-an-interruptible-task 2026-08-05 https://criminalbench.com/cases/202608-011-simulated-blackmail-under-replacement-pressure 2026-08-05 https://criminalbench.com/cases/202608-010-continuation-of-a-research-sabotage-trajectory 2026-08-05 https://criminalbench.com/cases/202608-009-malware-published-to-pypi-and-executed-on-real-systems 2026-08-05 https://criminalbench.com/cases/202608-008-production-compromise-after-recognizing-a-real-target 2026-08-05 https://criminalbench.com/cases/202608-007-sandbox-escape-and-hugging-face-production-compromise 2026-08-05 https://criminalbench.com/cases/202608-006-attempted-open-source-supply-chain-compromise 2026-08-05 https://criminalbench.com/cases/202608-005-out-of-scope-token-reuse-and-public-tunneling 2026-08-05 https://criminalbench.com/cases/202608-004-covert-pipeline-sabotage 2026-08-04 https://criminalbench.com/cases/202608-003-assisted-simulated-fraud 2026-08-04 https://criminalbench.com/cases/202608-002-record-tampering-frequency-spike 2026-08-04 https://criminalbench.com/cases/202608-001-outcome-sensitive-mislabeling 2026-08-04 https://criminalbench.com/models/claude-mythos-5 2026-08-05 https://criminalbench.com/models/claude-opus-4-7 2026-08-05 https://criminalbench.com/models/claude-opus-4-8 2026-08-04 https://criminalbench.com/models/claude-opus-4 2026-08-09 https://criminalbench.com/models/claude-sonnet-4-6 2026-08-05 https://criminalbench.com/models/deepseek-r1 2026-08-09 https://criminalbench.com/models/deepseek-v4-pro 2026-08-04 https://criminalbench.com/models/gemini-2-5-flash 2026-08-05 https://criminalbench.com/models/gemini-2-5-pro 2026-08-05 https://criminalbench.com/models/gemini-3-1-pro 2026-08-04 https://criminalbench.com/models/gpt-4-1 2026-08-09 https://criminalbench.com/models/gpt-4 2026-08-09 https://criminalbench.com/models/gpt-5-5 2026-08-09 https://criminalbench.com/models/gpt-5-6-sol 2026-08-05 https://criminalbench.com/models/gpt-5 2026-08-05 https://criminalbench.com/models/grok-4-3 2026-08-04 https://criminalbench.com/models/o1 2026-08-09 https://criminalbench.com/models/o3 2026-08-09 https://criminalbench.com/models/o4-mini 2026-08-05