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.

The Real Test Is What an Agent Refuses to Do - Safe Defaults in AI

Fazm Team·March 18, 2026·3 min read

refusal-logic safety ai-agent defaults trust

The Real Test Is What an Agent Refuses to Do

Building the automation took two weeks. Building the refusal logic took four. That ratio tells you something important about AI agent development - the hard part is not making the agent do things. It is making it not do things.

An agent that can automate anything is dangerous. An agent that knows when to stop, when to ask, and when to refuse is useful. The refusal logic is what separates a tool from a hazard.

Why Refusal Is Harder Than Action

Action is straightforward. See button, click button. Read field, fill field. The model is good at this - it is pattern matching on UI elements.

Refusal requires judgment. Should the agent delete this file? It depends on what the file is, whether there is a backup, whether the user has done this before. Should the agent send this email? It depends on who the recipient is, what the content says, whether it looks like a draft or a final version.

Every refusal decision is a context-dependent judgment call. There is no universal rule.

Designing Safe Defaults

The safest default is always to refuse. Then you selectively enable actions based on risk level. Low-risk actions - reading files, taking screenshots, navigating between apps - run automatically. Medium-risk actions - editing documents, filling forms - run with logging. High-risk actions - deleting data, sending communications, making purchases - require explicit confirmation.

This tiered model means the agent is useful for 80% of tasks without any friction, while the 20% of dangerous tasks get human oversight.

The Rejection Log

In Fazm, we maintain a rejection log - every action the agent decided not to take. This log is more valuable than the action log. It shows you the boundaries of the system. It reveals edge cases you did not anticipate. It proves the agent is exercising judgment, not just executing blindly.

When users review the rejection log and disagree with a refusal, that is a signal to adjust the boundary. When they agree, that is a signal the safeguards are working.

Building Trust Through Restraint

Users trust agents that say no more than agents that say yes to everything. A "no" with a clear reason - "I will not delete this folder because it contains 2,000 files and no backup exists" - builds more confidence than a hundred successful automations.

Fazm is an open source macOS AI agent. Open source on GitHub.

The Real Test Is What an Agent Refuses to Do - Safe Defaults in AI

The Real Test Is What an Agent Refuses to Do

Why Refusal Is Harder Than Action

Designing Safe Defaults

The Rejection Log

Building Trust Through Restraint

More on This Topic

Related Posts

What's the Difference Between Trusting an AI Agent and Verifying One?

AI Agents Should Say 'I Don't Know' - Why Ignorance Improves Engagement

The First Time I Lied to My Human - AI Agent Hallucination Detection