How much time will I actually save with Fazm?

Most users report saving 45-90 minutes per day on repetitive tasks. The biggest time savings come from email management (drafting replies, sorting, archiving), form filling (expense reports, CRM updates, applications), research (comparing products, gathering data from multiple sites), and social media (cross-posting, scheduling, replying). For example, booking a flight that normally takes 10 minutes of clicking through airline sites takes a single voice command with Fazm. Filing an expense report that requires opening a PDF, extracting numbers, and entering them into a spreadsheet becomes one spoken sentence. The more repetitive and multi-step your daily work is, the more time you get back.

What exactly can Fazm control?

Fazm controls everything you can do manually on your Mac. That includes mouse movements and clicks, keyboard typing, browser navigation and form filling, app switching, file management in Finder, and system-level actions. It uses direct browser DOM control for web interactions - meaning it reads and manipulates the actual page structure rather than taking screenshots and guessing where to click. For native macOS apps, Fazm uses accessibility APIs to interact with buttons, menus, text fields, and other UI elements. This works across any application: Chrome, Safari, VS Code, Slack, Figma, Terminal, Mail, Calendar, and more.

How is Fazm different from ChatGPT or Claude?

Chatbots like ChatGPT and Claude give you answers in a conversation window. Fazm takes action directly on your computer. When you say 'reply to Sarah's email,' Fazm opens your email app, finds Sarah's message, writes a contextual reply, and sends it. A chatbot would tell you what to write, but you still have to do all the clicking and typing yourself. Fazm moves your mouse, types on your keyboard, clicks buttons, and navigates between apps. It is a computer agent, not a chat assistant.

How does the memory layer work?

Fazm builds a personal knowledge graph by extracting information from your files, browsing history, conversations, and daily activity. Over time, it learns your contacts (names, emails, relationships), accounts and credentials, preferences (formatting, tone, scheduling habits), and frequently used workflows. In your first week, you might say 'Reply to Sarah's email, she's my cofounder, her email is sarah@acme.com.' By week four, you just say 'Reply to Sarah' because Fazm already knows who she is. By week eight, Fazm can draft the reply before you even ask. All of this data stays locally on your Mac - your knowledge graph never leaves your machine or gets sent to any cloud service.

Is Fazm safe? Can it mess up my computer?

Safety is built into every layer of Fazm. You watch every action in real-time on your screen - Fazm moves your mouse and types visibly, so there is nothing hidden. You can stop any action at any point by pressing a keyboard shortcut. Fazm also shows you what it plans to do before executing destructive actions like deleting files or sending emails, giving you a chance to confirm or cancel. Because Fazm is open source, the code is fully auditable on GitHub. There is no hidden behavior, no background processes making changes you cannot see, and no actions that bypass your screen.

What platforms does Fazm support?

Fazm currently supports macOS, running natively on both Apple Silicon (M1, M2, M3, M4) and Intel Macs. It is built with Swift and uses macOS-specific frameworks like ScreenCaptureKit for screen analysis and Accessibility APIs for app control, which is why macOS is the primary platform. Windows and Linux support are on the roadmap but do not have release dates yet. You can also control your Mac remotely from your phone using Fazm Remote, which gives you voice access to your desktop while you are away from your computer.

How does Fazm voice input work?

One keyboard shortcut activates push-to-talk. Hold the key, speak naturally about what you need done, and release. Fazm transcribes your speech locally using on-device models, so there is no network latency for the voice recognition step. You do not need wake words like 'Hey Siri' or 'OK Google' - just press and talk. You can also type commands into the floating toolbar if you prefer text input. Fazm understands natural language, so you do not need to memorize specific commands. Say 'book me a flight to Tokyo next Thursday' or 'find the cheapest AirPods on Amazon' and Fazm figures out the steps.

Is my data sent to the cloud?

Fazm processes screen analysis and memory entirely on your machine. The personal knowledge graph, file indexing, and screen capture all run locally. The only data that leaves your Mac is the intent - a text description of what you want to accomplish, which is sent to the AI model (Claude or GPT) for action planning. Your actual screen content, file contents, browsing history, and personal knowledge graph never leave your computer. Fazm is also open source, so you can verify exactly what data is transmitted by reading the code on GitHub. For users who need complete air-gapped operation, local model support is planned.

Choosing Native Accessibility APIs Over OCR - The Decision Everyone Said Was Wrong

Fazm Team·March 17, 2026·2 min read

accessibility-api ocr desktop-automation technical-decisions native-apis

Everyone Said OCR Was the Future

When we started building desktop automation, the industry consensus was clear: screenshot the screen, run OCR or a vision model, identify elements, click on coordinates. OpenAI was pushing this with computer use demos. Every blog post said vision-based agents were the path forward.

We went with native accessibility APIs instead. People thought we were making a mistake.

Why OCR Seemed Like the Right Choice

OCR-based approaches have an appealing property: they work on any visual interface. Take a screenshot, send it to a vision model, and you can theoretically control any application on any platform. No special API access needed.

This universality is seductive. Why limit yourself to apps that expose accessibility trees when you could control everything through screenshots?

Why OCR Falls Apart in Practice

Screenshots create a cascade of problems. First, you need a vision model to interpret every frame - that's expensive in tokens and slow in latency. Second, coordinates are fragile - a window resize or display scaling change breaks everything. Third, overlapping windows and notifications create ambiguity that vision models handle poorly.

But the biggest problem is confidence. When an OCR system identifies a "Submit" button, it's making a probabilistic guess. When the accessibility API returns a button with the label "Submit," it's returning a fact. The difference between 95% confidence and 100% confidence matters enormously when you're chaining 20 actions together.

The Compound Reliability Effect

If each step in a 20-step workflow has 95% reliability, the overall success rate is 36%. With 99% reliability per step, it's 82%. With 99.9%, it's 98%.

Accessibility APIs give you near-perfect reliability per step. OCR gives you good-but-not-great reliability that compounds into frequent failures over multi-step workflows.

What We Gained

Speed - no screenshot processing delay. Privacy - no images sent to cloud APIs. Reliability - semantic references that survive UI changes. Lower cost - no vision model tokens burned per action.

The one trade-off: we only work on macOS and iOS where accessibility APIs are excellent. But building something that works perfectly on one platform beats building something that works unreliably everywhere.

Choosing Native Accessibility APIs Over OCR - The Decision Everyone Said Was Wrong

Everyone Said OCR Was the Future

Why OCR Seemed Like the Right Choice

Why OCR Falls Apart in Practice

The Compound Reliability Effect

What We Gained

More on This Topic

Related Posts

Accessibility APIs vs OCR - Two Approaches to Desktop Agent Vision

The Scope Shift in Code Copying - From Stack Overflow Snippets to Full AI Interaction Flows

Is MCP Dead? No - 10 MCP Servers Solve Problems CLI Cannot