Everyday AI Podcast – An AI and ChatGPT Podcast

Ep 877: ChatGPT Voice is Like Jarvis: How to use the New Feature and the 7 biggest unlocks (Replay)

October 9, 2026

AI Summary

5 min read

ChatGPT Voice Is Like Having Jarvis

Jordan Wilson, host of the Everyday AI Podcast, calls ChatGPT Voice "the most criminally underhyped release in AI ever." After spending an hour walking around Chicago with an AirPod in his ear, controlling his home computer from his iPhone, he says the feature has changed not just how he works, but how he thinks about work. The key insight: this isn't dictation, and it isn't the advanced voice mode that lets you have a natural conversation. This is a voice-controlled agent that can open programs, browse the web, run code, access your connected tools, and execute multi-step tasks — all while you ramble instructions hands-free.

What Makes ChatGPT Voice Different

The confusion starts with naming. There are four distinct ways to use your voice with ChatGPT, and only one of them does real work.

Dictation is the oldest — click the microphone icon and your speech becomes text. Advanced voice mode (GPT Live) is the full-duplex conversational layer that can talk and listen simultaneously, interrupt and be interrupted, and even browse the web or run code in the background. But it cannot access your connected tools, your files, your email, or your computer.

Continue reading the full summary in the app — free to try.

Read Full Summary →

Free • No credit card required

Listen to Audio Summary Open in App

Never miss an episode of Everyday AI Podcast – An AI and ChatGPT Podcast

Get every new episode summarized in your inbox — free, ~5 minutes to read.

No spam. Unsubscribe anytime.

What you'll learn

1 (00:01) **ChatGPT Voice as Your Own Jarvis** - The host declares this feature a rare "how I think about work" shift, comparing it to Iron Man's Jarvis for hands-free computer control.
2 (02:07) **What ChatGPT Voice Actually Is** - Clarifying the confusing launch: it's the GPT Live voice model combined with the agentic execution layer inside the desktop app (ChatGPT Work / Codex).
3 (04:50) **The Four Phases of Voice Interaction** - A taxonomy to cut through confusion: dictation, old voice mode, GPT Live, and the new ChatGPT Voice inside Codex.
4 (08:14) **Why Connectors Are the Missing Piece** - The key difference: GPT Live can't use your integrations, but ChatGPT Voice inside Codex can access all your connected tools and MCP servers.
5 (09:18) **How to Start a Voice Chat** - A quick navigation guide to the correct button, which is easy to confuse with the dictation microphone.
6 (11:36) **Advanced Tip #1: Any New Chat Can Access All Threads** - A major workflow unlock: you cannot resume an old thread with voice, but a new chat can read, write, and send messages to any existing thread.
7 (16:07) **Advanced Tip #2: Connectors Are the Biggest Unlock** - The host argues this is the most under-discussed advantage: the ability to triage email, Slack, CRM, and other tools entirely by voice.

+ Full timestamped outline available in the app

Guests on this episode

Jordan Wilson

Show Notes

People are calling the new ChatGPT Voice their “AGI moment.” 🤖

After testing it over the past week, I understand why.

It seems like the missing link between the ever-capable desktop agent and the human worker not having to type in front of said desktop all day.

It’s less like “voice mode” and more like having your own JARVIS.

After testing thousands of AI tools, this has instantly become one of the most useful features I’ve ever used.

Ep 877: ChatGPT Voice is Like Jarvis: How to use the New Feature and the 7 biggest unlocks (Replay)

Newsletter: Sign up for our free daily newsletter
More on this Episode: Episode Page
Today's Episode on LinkedIn: Thoughts on this? Join the convo on LinkedIn and connect with other AI leaders.

Upcoming Episodes: Check out the upcoming Everyday AI Livestream lineup
Website: YourEverydayAI.com
Email The Show: [email protected]
Connect with Jordan on LinkedIn

Topics Covered in This Episode:

ChatGPT Voice as Jarvis-Style AI Agent
ChatGPT Voice vs. Old Voice Modes
Full Duplex GPT Live Model Explained
Desktop ChatGPT Voice: Codex Integration
Advanced ChatGPT Voice Workflow Use Cases
Utilizing Connectors and MCP Integrations
Stacking and Scheduling Tasks with Voice
Hands-Free Computer Control via iPhone
Durable Written Artifacts and Accountability
Omnipresent Desktop Orb Agent Features
Live Demo: Real-World Voice Command Tasks
Productivity Gains from Voice-Driven Automation

Timestamps:

00:00 Experiencing the new ChatGPT voice

04:30 AI demonstration details

08:44 Accessing GPT data sources

11:00 Differentiating Codex from ChatGPT Voice

16:08 Limitations of GPT Live integrations

19:00 Automating Tasks with Voice Commands

21:47 Using ChatGPT voice on iPhone

25:09 Using AirPods with your Mac

27:09 Starting a Chat GPT Demo

31:21 Launching the Parrot Program

35:47 Demoing JGPT's capabilities

39:18 Demonstrating multitasking techniques

40:58 Wor