Article · 2026-07-22 · 6 moments

Computer and browser use in Codex (5 real examples)

Watch now | 🎙️ I show you exactly how I use browser use and computer use in Codex to QA my app, manage LinkedIn, and shop for Hawaii, including the under-prompting trick that makes frontier models work harder ✦ AI generated

Read original ↗

GUEST Claire Vo

timeline · colored by role

Claim◆

Browser use and computer use via Codex (ChatGPT desktop app) is a daily driver for both personal and professional workflows, not just a novelty feature.

The speaker positions Codex browser and computer use as core daily infrastructure rather than an occasional experiment.

transcript

Claire Vo: Today I'm walking you through one of my absolute favorite AI features right now: browser and computer use via Codex (the ChatGPT desktop app). I use this every single day, personally and professionally, and I wanted to share the specific workflows I've built, the moments that surprised me, and the mental model that makes it actually click.

supports · 1

Example

Codex performed exhaustive QA testing on the onboarding flow, surfacing 11 issues including one high-severity blocker, with a Google Sheet of screenshots.

Codex did exhaustive automated QA testing that a human would skip, catching real bugs.

transcript

Claire Vo: How I use Codex to QA my onboarding flow, including exhaustive mobile testing I would never do manually... Results: 11 issues, one high-severity blocker, one Google Sheet with screenshots

gives example · 2

Mechanism

EJ Lawless's persona-impersonation trick allows Codex to simulate different user perspectives, surfacing friction points the builder cannot see.

A technique where the AI adopts a specific user persona to stress-test a product, revealing blind spots the creator cannot see.

transcript

Claire Vo: How my husband EJ Lawless's persona-impersonation trick surfaces friction points I can't see as the builder

extends · 1gives example · 2

Example

Codex browser use handles real-world personal tasks autonomously: clearing a LinkedIn inbox, shopping Free People for breastfeeding-friendly Hawaii clothes, and managing a router from out of state.

Practical personal-use examples showing the range of autonomous tasks Codex handles across professional and personal domains.

transcript

Claire Vo: How I use browser use to get through my LinkedIn inbox without touching it myself... How I had Codex shop Free People's sale and add 10 medium-size items to my cart (breastfeeding-friendly and Hawaii-ready)... Three more computer-use shortcuts: filling annoying forms, creating Google Sheets mid-workflow, and managing a router

gives example · 4

Mechanism

Computer use can control iPhone mirroring, enabling a Mac to operate an iPhone and extending automation to mobile workflows.

Chaining Codex computer use with Apple's iPhone mirroring bridges desktop AI automation and mobile device control.

transcript

Claire Vo: How computer use can control iPhone mirroring so your Mac can technically operate your phone

extends · 2

Mechanism◆

Under-prompting frontier models — giving goals rather than detailed step-by-step instructions — produces better results because it forces the model to reason through the problem rather than mechanistically follow a script.

The counterintuitive insight that less-detailed prompts yield superior outcomes with advanced models by compelling genuine reasoning.

transcript

Claire Vo: Why under-prompting frontier models gets better results than detailed step-by-step instructions

extends · 1

Highlight slides

Codex Browser & Computer Use: Daily Infrastructure✦ from: Browser use and computer use via Codex (ChatGPT desktop app) is a daily driver for both personal and professional workflows, not just a novelty feature.What Makes It Click✦ from: Browser use and computer use via Codex (ChatGPT desktop app) is a daily driver for both personal and professional workflows, not just a novelty feature.The Counterintuitive Power of Under-Prompting✦ from: Under-prompting frontier models — giving goals rather than detailed step-by-step instructions — produces better results because it forces the model to reason through the problem rather than mechanistically follow a script.Why Less Detail Yields Better Results✦ from: Under-prompting frontier models — giving goals rather than detailed step-by-step instructions — produces better results because it forces the model to reason through the problem rather than mechanistically follow a script.

Related episodes

Video · BG2 Pod

The SpaceX IPO, Fable 5, AI Capex Update & Market Check w/ Gavin Baker, Andrew Fox & Clark Tang

→

Article · ByteByteGo Newsletter

AI Customer Support at Scale: The Travel Industry’s $Billion Bet

→

Article · Simon Willison's Weblog

simonw/pedalican

→