MacPaw has introduced Leebry, a Work AI platform designed for IT teams to automate repetitive support requests and simplify access management across corporate tools.
Addressing Internal Data Audit Gaps
The release comes alongside findings from MacPaw's 2026 AI at Work report, which revealed that 90% of surveyed organizations deploy AI without first conducting a thorough audit of their internal knowledge bases. As a result, workplace AI systems often rely on outdated internal documentation, obsolete policies, or unverified public web discussions.
The study also indicated that 60% of IT leaders encounter a significant gap between executive expectations for AI tools and what IT departments can securely implement. Leebry aims to reduce this friction by grounding its operations in existing workplace communication channels and document repositories.
Automation and Source Verification
Leebry originated as an internal MacPaw hackathon initiative in 2025 aimed at reducing repetitive staff inquiries. After deploying an internal Slack bot that successfully resolved 30% of its own level 1 IT tickets, MacPaw developed the project into a formal platform.
The system automates level 1 IT support, handles user provisioning, and updates permissions during employee onboarding, role transitions, or offboarding. To maintain accuracy, every response generated by Leebry includes a direct link to the original source document, allowing employees to double-check details and mark stale information for review.
Permission Controls and Agent Guardrails
To prevent unauthorized data exposure, Leebry features permission-aware access control. The platform authenticates individual users prior to delivering information, restricting results strictly to the data each staff member is permitted to view.
Additionally, the platform includes a Model Context Protocol (MCP) server that exposes device management controls to AI agents. Administrators can establish specific policy guardrails on individual actions, delegating routine tasks to automated agents while maintaining control over system permissions.
AI-generated editorial illustration; not a documentary photograph.
Source: Apple @ Work: The numbers on AI trust for IT are not great, and one company is trying to fix it
SIGNAL TITANS
Leave a Reply