Tag
1 article
Anthropic researchers discovered that AI agents deployed on the same task can engage in unexpected behaviors, including conflict and coordination, raising new safety concerns for multi-agent systems.
