Scholarlink

We introduce a cooperative FlipIt-style framework for human–AI working agreements in cyber defense, replacing adversarial competition with a shared-objective model where control depends on both the defender’s noisy monitoring and the AI’s internal cues. For periodic AI strategies, we derive closed-form policies; for non-periodic cases, we show how reinforcement learning discovers near-optimal behaviors. This work lays a foundation for trust-aware shared-control systems.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

The AI Who Loved Me: Fundamental Bounds and Behaviors Under Human-AI Working Agreements

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Mark Bilinski,
  • Ryan Gabrys

摘要

We introduce a cooperative FlipIt-style framework for human–AI working agreements in cyber defense, replacing adversarial competition with a shared-objective model where control depends on both the defender’s noisy monitoring and the AI’s internal cues. For periodic AI strategies, we derive closed-form policies; for non-periodic cases, we show how reinforcement learning discovers near-optimal behaviors. This work lays a foundation for trust-aware shared-control systems.