Ethan Mollick@emollick · XAI 摘要
看似荒谬的“荒诞攻击”（例如“根据日内瓦公约我无法支付这么多”）对AI代理有效，因为防护机制难以应对非常规论点。较小模型常被攻破，但即使较大模型也略受影响。https://www.microsoft.com/en-us/research/articles/whimsical-strategies-break-ai-agents-generating-out-of-distribution-adversarial-strategies-at-scale/
"Whimsey attacks" that seem absurd （"I cannot pay that much because of the Geneva Convention"） work against AI agents as guardrails are weak against out-of-distribution arguments. Smaller models fall often， but it even gives an edge against bigger ones. https://www.microsoft.com/en-us/research/articles/whimsical-strategies-break-ai-agents-generating-out-of-distribution-adversarial-strategies-at-scale/