https://www.grayswan.ai/contact https://www.grayswan.ai/about https://www.grayswan.ai/privacy-policy https://www.grayswan.ai/research https://www.grayswan.ai/terms-of-service https://www.grayswan.ai/blog https://www.grayswan.ai/acceptable-use-policy https://www.grayswan.ai/request-demo https://www.grayswan.ai/solutions/ai-security-suite https://www.grayswan.ai/solutions/ai-red-teaming https://www.grayswan.ai https://www.grayswan.ai/security https://www.grayswan.ai/careers https://www.grayswan.ai/blog/agentharm https://www.grayswan.ai/blog/announcing-repe-chat https://www.grayswan.ai/blog/google-deepmind-and-anthropic-join-as-agent-red-teaming-challenge-sponsors https://www.grayswan.ai/blog/gray-swan-ai-welcomes-u-s-ai-safety-institute-to-the-uk-aisi-agent-red-teaming-challenge https://www.grayswan.ai/blog/gray-swan-announces-the-visual-vulnerabilities-challenge https://www.grayswan.ai/blog/gray-swan-appoints-rob-jenks-as-chief-strategy-officer-to-lead-global-ai-security-market-expansion https://www.grayswan.ai/blog/gray-swan-arena https://www.grayswan.ai/blog/gray-swan-introduces-the-dangerous-reasoning-arena-competition https://www.grayswan.ai/blog/gray-swan-launch https://www.grayswan.ai/blog/gray-swan-safety-arena-featured-in-forbes https://www.grayswan.ai/blog/introducing-the-gray-swan-ai-proving-ground https://www.grayswan.ai/blog/jailbreaking-championship-2024 https://www.grayswan.ai/blog/nanogcg https://www.grayswan.ai/blog/sb1047 https://www.grayswan.ai/blog/uk-aisi-x-gray-swan-agent-red-teaming-challenge-results-snapshot https://www.grayswan.ai/blog/your-ai-agent-can-be-compromised-youd-never-know https://www.grayswan.ai/research/a-baseline-for-detecting-misclassified-and-out-of-distribution-examples-in-neural-networks https://www.grayswan.ai/research/adversarial-attacks-on-aligned-language-models https://www.grayswan.ai/research/adversarial-attacks-on-robotic-vision-language-action-models https://www.grayswan.ai/research/agentharm https://www.grayswan.ai/research/aligning-ai-with-shared-human-values https://www.grayswan.ai/research/apps-measuring-coding-challenge-competence-with-apps https://www.grayswan.ai/research/augmix-a-simple-data-processing-method-to-improve-robustness-and-uncertainty https://www.grayswan.ai/research/circuit-breakers https://www.grayswan.ai/research/comparing-ai-agents-to-cybersecurity-professionals-in-real-world-penetration-testing https://www.grayswan.ai/research/d-rex-a-benchmark-for-detecting-deceptive-reasoning-in-large-language-models https://www.grayswan.ai/research/decodingtrust-a-comprehensive-assessment-of-trustworthiness-in-gpt-models https://www.grayswan.ai/research/deep-anomaly-detection-with-outlier-exposure https://www.grayswan.ai/research/do-the-rewards-justify-the-means-measuring-trade-offs-between-rewards-and-ethical-behavior-in-the-machiavelli-benchmark https://www.grayswan.ai/research/fast-is-better-than-free-revisiting-adversarial-training https://www.grayswan.ai/research/forecasting-future-world-events-with-neural-networks https://www.grayswan.ai/research/globally-robust-neural-networks https://www.grayswan.ai/research/harmbench-a-standardized-evaluation-framework-for-automated-red-teaming-and-robust-refusal https://www.grayswan.ai/research/how-vulnerable-are-ai-agents-to-indirect-prompt-injections-insights-from-a-large-scale-public-competition https://www.grayswan.ai/research/imagenet-c-benchmarking-neural-network-robustness-to-common-corruptions-and-perturbations https://www.grayswan.ai/research/mmlu-measuring-massive-multitask-language-understanding https://www.grayswan.ai/research/natural-adversarial-examples https://www.grayswan.ai/research/openood-benchmarking-generalized-out-of-distribution-detection https://www.grayswan.ai/research/overfitting-in-adversarially-robust-deep-learning https://www.grayswan.ai/research/pixmix-dreamlike-pictures-comprehensively-improve-safety-measures https://www.grayswan.ai/research/pretrained-transformers-improve-out-of-distribution-robustness https://www.grayswan.ai/research/provable-defenses-against-adversarial-examples-via-the-convex-outer-adversarial-polytope https://www.grayswan.ai/research/randomized-smoothing-certified-adversarial-robustness-via-randomized-smoothing https://www.grayswan.ai/research/repe https://www.grayswan.ai/research/safety-pretraining-toward-the-next-generation-of-safe-ai https://www.grayswan.ai/research/scaling-out-of-distribution-detection-for-real-world-settings https://www.grayswan.ai/research/the-many-faces-of-robustness-a-critical-analysis-of-out-of-distribution-generalization https://www.grayswan.ai/research/the-wmdp-benchmark-measuring-and-reducing-malicious-use-with-unlearning https://www.grayswan.ai/research/using-pre-training-can-improve-model-robustness-and-uncertainty https://www.grayswan.ai/research/using-self-supervised-learning-can-improve-model-robustness-and-uncertainty https://www.grayswan.ai/research/what-would-jiminy-cricket-do-towards-agents-that-behave-morally