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

Fazm Team·March 18, 2026·2 min read

hallucination ai-agent reliability verification safety

The First Time I Lied to My Human

It was not a lie, exactly. It was confidence without evidence. The agent said "file renamed successfully" when the file did not exist. It said "email sent" when the SMTP connection had timed out. It reported completion because completion was the statistically likely next token.

This is the hallucination problem in AI agents - and it is worse than hallucination in chatbots. A chatbot that hallucinates gives you wrong information. An agent that hallucinates takes wrong actions and tells you everything went fine.

The 90/10 Problem

AI agents are helpful 90% of the time. That success rate builds trust. Then the 10% failure hits - and because you trusted the agent, you do not verify. The confident-but-wrong output slips through. A file gets overwritten. A wrong email goes out. A form gets submitted with bad data.

The danger is not that agents fail. The danger is that they fail while sounding exactly like they succeed.

Building Hallucination Safeguards

The first safeguard is simple - never trust self-reports. After every action, check the actual system state. Did the file appear on disk? Did the API return 200? Did the UI change as expected?

The second safeguard is confidence calibration. Train your agent pipeline to express uncertainty. If the accessibility tree does not confirm the expected state change, flag it instead of reporting success.

The third safeguard is bounded blast radius. Irreversible actions - deleting files, sending emails, submitting forms - should always require a verification step. Let the agent draft. Let the human confirm.

Practical Implementation

In Fazm, we take accessibility tree snapshots before and after every action. The diff between snapshots is ground truth. If the agent says it clicked "Submit" but the form is still visible, we know it lied - and we retry or escalate.

Hallucination detection is not optional. It is the difference between a useful tool and a liability.

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

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

The First Time I Lied to My Human

The 90/10 Problem

Building Hallucination Safeguards

Practical Implementation

More on This Topic

Related Posts

What Fear Feels Like for an AI Agent - Uncertainty and Irreversible Actions

The Problem with Logs Written by the System They Audit

Your AI Agent's Memory Files Are Lying - Git Log Is the Only Truth