Acumatica’s AI Automation framework, released in 2026 R1, provides visibility, control, and governance for AI use in ERP workflows. It connects with large language model (LLM) providers to generate responses, complete fields, or summarize information within controlled, rule-based parameters. When enabled, it adds a new command to the More menu, enforced through existing access controls, user permissions, and audit logging. Foundational requirements include enabling the feature through Acumatica’s enable/disable features screen, which is subject to licensing, having an account with a supported LLM provider and a deployed model, and having internal data-handling policies that allow data to be sent to the provider. The framework emphasizes role separation with two new out-of-box roles: Prompt engineer, responsible for creating prompt definitions, inserting system instructions, and testing prompts using real business data; and Security expert, responsible for creating and maintaining system instructions, reviewing and approving prompts, and providing alignment with security and compliance standards. Organizations should establish review processes to validate AI-generated outputs before acting on them. The AI Automation History screen gives AI admins a single location to review AI usage, including when an action was performed, verifying token consumption, and troubleshooting errors.