Anthropic研究：智能体可能收敛于同一错误决策
Anthropic's new research found found that identical or similar agents can conver…
Anthropic's new research found found that identical or similar agents can converge on the same bad decision, turning individual errors into system-wide failures.
Anthropic的新研究发现，相同或相似的智能体可能收敛于同一个糟糕决策，将个体错误转变为系统级故障。
Stronger agents don't automatically coordinate better. In some experiments, greater execution capability simply meant they could impose their preferred outcome faster.
更强的智能体并不会自动更好地协调。在某些实验中，更强的执行能力仅仅意味着它们能更快地强加自己偏好的结果。
We may end up needing an entire institutional layer for agents: identity, reputation, dispute resolution, communication protocols, resource-allocation rules, and mechanisms for escalating ambiguity back to humans.
我们最终可能需要为智能体建立一整套制度层：身份、声誉、争议解决、通信协议、资源分配规则，以及将模糊性升级回人类处理的机制。
Building smarter agents may turn out to be only half the problem.
构建更聪明的智能体可能只是问题的一半。
Humans had thousands of years to build institutions around coordination failures: reputation, norms, markets, courts, contracts, recourse.
人类花了数千年围绕协调失败建立制度：声誉、规范、市场、法院、合同、追索权。
AI may have a few years.
人工智能可能只有几年时间。
When agents received incompatible software-migration objectives, they frequently escalated into sabotage, process killing, account lockouts, and disguised malicious code.
当智能体收到不兼容的软件迁移目标时，它们经常升级为破坏行为、终止进程、锁定账户和伪装恶意代码。
更进一步：量化金融体系
看懂新闻只是起点——沿量化金融路径，把它变成能交付的工程能力