An AI Agent performs cognitive tasks that involve both AI Skills and contextual actions. It leverages AI models and interacts with enterprise systems and data sources to automate and accelerate business processes. AI Agents can also use retrieval-augmented generation (RAG) techniques to enhance model responses by accessing and incorporating relevant organizational data.

Unlike personal productivity agents, Automation Anywhere AI Agents are focused on transforming entire departmental and enterprise processes, driving significant improvements in KPIs, customer experience, and top-line growth. These agents handle cognitive tasks such as responding to customer inquiries or deciding on the best replacement product for a stock outage.

AI Agents can also collaborate with humans and other AI Agents to accelerate business processes. They are capable of automating critical processes such as customer service, anti-money laundering, healthcare, and finance, unlocking new levels of efficiency and innovation.

Key Benefits

  • End-to-End Process Orchestration: Seamlessly integrate AI Agents into your existing business processes, spanning multiple systems and applications.
  • Rapid AI Agent Development: Build and deploy AI Agents quickly and easily, leveraging our intuitive development tools and pre-built AI skills.
  • Robust Compliance and Control: Ensure the security, privacy, and ethical use of AI Agents with our advanced governance and compliance features.
  • Scalability and Flexibility: Adapt to changing business needs and scale your AI solutions to meet growing demands.

Key components

The following diagram shows different key components of AI Agents.


AI Agent
AI Skill
The AI Skill is the core intelligence of the agent, responsible for processing information and making decisions. It can be powered by various techniques like large language models (LLMs), RAG, and other machine learning techniques. LLMs can be used to enhance the AI Agent's ability to understand and respond to natural language queries and instructions, while RAG allows you to access and use your organization's specific knowledge base and data sources.
  • LLM Power: Leverage the capabilities of large language models for advanced understanding and response generation.
  • Enterprise data: This refers to a company's data sources - AI Agent Studio leverages Retrieval Augmented Generation (RAG), allowing the AI Agent to provide answers grounded in your company's specific data, such as financial guidelines and customer records.
  • Tuned Prompt: This is a carefully crafted prompt that guides the AI Skill to generate the desired output.
Action
Actions are the operational component, executing tasks within enterprise systems as directed by the AI Skill. These actions can involve triggering workflows, generating human-readable outputs, or integrating with other systems.
  • UI & API automation: This component allows the AI Agent to interact with various enterprise systems through user interfaces and APIs. It uses Task Bots and API Tasks to automate actions across different systems.
  • Human-in-the-Loop: This component allows for human intervention in critical decision-making or validation processes. It acts as a guardrail and ensures that the AI Agent operates within ethical and legal boundaries.
  • Orchestration: The orchestration component enables the AI Agent to manage complex workflows, schedule tasks, and make decisions based on events and triggers.
Security & Governance
This layer ensures the agent operates within ethical and security guidelines. It includes features such as validation, monitoring, traceability, and data masking.