Atlassian lays groundwork for humans and AI agents to work side by side

Atlassian announced a platform update called Agentic Multiplayer Protocol (AMP) to change how humans and artificial intelligence agents collaborate and work together. The update aims to bring "multiplayer" into a fully governed system of work, allowing humans and agents to operate side by side in the open. Atlassian's Head of AI Product, Jamil Valliani, emphasized that humans and agents are working together in a dynamic space, with limitations in place to ensure safety.

To ensure agents have the necessary data context, Atlassian introduced AMP, which assigns each agent an identity assigned by administration, including authority and scope. Valliani highlighted that while agents are autonomous, they cannot operate without teamwork. To tackle this, Atlassian introduced Rovo Work, a new mode for Rovo Chat that handles complex, multi-step tasks that humans review and approve.

Rovo Work allows agents to go beyond straightforward work completion. For example, when asked to create an Instagram-ready reel, the agent will research the required instructions, learn what it needs for the output format, and obtain tooling for video synthesis, even if it doesn't have this information in its model training data. This approach differs significantly from standard AI models, which rely on predefined data and struggle to learn new tasks.

Atlassian also introduced Loom, a product that allows users to record videos of themselves and their computer screens to provide instructions for agents. This feature bridges the gap between human and agent collaboration by enabling users to visually point out what they need, alongside verbal instructions. For developers, Atlassian's broader platform changes provide an expanded context and operating environment for AI agents, with Rovo Code Search bringing source code into the Teamwork Graph, a centralized data intelligence and context engine.