IrisGo, a startup backed by Andrew Ng, looks to become the AI desktop buddy you never knew you needed

2026-05-20 · Source: AI News & Artificial Intelligence | TechCrunch · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Software Development & Engineering, Robotics & Autonomous Systems · Depth: Fundamental Awareness, short

Summary

IrisGo, a startup backed by Andrew Ng's AI Fund, recently closed a \$2.8 million seed round to develop a proactive AI desktop companion for PCs. This application learns user workflows and automates repetitive tasks with minimal prompting, aiming to shift knowledge workers from manual office work to autonomous workflows. Co-founded by former Apple engineer Jeffrey Lai, IrisGo features a "skills" library for tasks like email drafting and invoice processing, alongside a coding assistant. A key differentiator is its hybrid architecture, processing much data on-device for enhanced privacy, with cloud processing requiring explicit user authorization and end-to-end encryption. IrisGo has launched beta versions for macOS and Windows and secured a pre-installation deal with Acer, with plans for further partnerships.

Key takeaway

For Directors of AI/ML evaluating solutions for knowledge worker productivity, IrisGo's proactive AI desktop companion offers a path to automate repetitive office tasks. Its on-device processing architecture provides stronger privacy assurances than cloud-heavy alternatives, which is crucial for sensitive business data. Consider piloting IrisGo to reduce manual effort in workflows like invoice processing or report building, allowing your teams to focus on higher-level conceptual work.

Key insights

Proactive AI agents automate desktop workflows by learning user behavior and anticipating needs.

Principles

• Automate repetitive tasks for knowledge workers.
• Prioritize on-device processing for privacy.
• Learn from user behavior for task automation.

Method

The system records user steps for a task, such as an online coffee order, then stores and repeats that process for future automated execution without further instructions.

In practice

• Train an agent by demonstrating a task once.
• Utilize built-in skills for common office workflows.
• Explore pre-installation deals for wider adoption.

Topics

• Proactive AI
• Desktop Automation
• Workflow Automation
• On-device AI
• Data Privacy
• AI Agents

Best for: Investor, Executive, AI Product Manager, Director of AI/ML, Consultant, Software Engineer

Related on AIssential

• Grab my Ultimate Guide to Codex and catch up to the 1 in 1,600 people using it every week (mostly no code!) — Codex transforms computer interaction by integrating deeply across systems, enabling automation of complex, multi-step jobs beyond basic chat.
• Google turns Chrome into an AI coworker for the workplace — Google is integrating Gemini AI into Chrome for enterprise users to automate web tasks while enhancing security against unsanctioned AI.
• Put Claude to work on your computer — Claude can now directly control your computer, performing tasks via screen interaction when direct integrations are absent.
• 10 Everyday Tasks You Can Automate with AI Today (With n8n Templates) — Low-code platforms like n8n enable AI automation for common tasks, reducing manual effort and coding.
• This Week in AI — AI tools are rapidly integrating into daily workflows, offering enhanced automation, privacy, and multimodal capabilities.
• Notion just turned its workspace into a hub for AI agents — Notion's new developer platform integrates custom code, external data, and AI agents to create an orchestration hub.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by AI News & Artificial Intelligence | TechCrunch.