ChatGPT Codex Voice + browser + Sites: an expert’s AI workflow | Nick Baumann (OpenAI) · How I AI summary

Key takeaways

Recent AI tools like ChatGPT Codex and ChatGPT Work are becoming more powerful and accessible, enabling complex workflows through simple voice commands and automation.
The new voice interface with screen reading and task management drastically expands the potential use cases, making AI assistants more natural and manageable for everyday tasks.
Automation features like Heartbeats allow continuous monitoring and proactive notifications, effectively turning AI into a personal assistant that operates in background.
Building and deploying websites via ChatGPT Sites now enables anyone to create and share sophisticated online content easily, with customizable privacy and design options.
AI-driven content creation and editing, especially for video, is rapidly evolving, significantly lowering the barrier for creators and solo entrepreneurs to produce high-quality media.
Reducing latency in AI interactions is crucial for more natural, flowing conversations and workflows, and will likely be a major focus in future developments.
Voice interaction, while currently imperfect for office settings, offers a high bandwidth way to delegate tasks and manage workflows naturally, pointing towards future seamless integrations.

Summary

Introduction to Advanced AI Capabilities with Nick Baumann

The podcast begins with an introduction to Nick Baumann from OpenAI, highlighting his role in building and testing advanced tools like ChatGPT Codex and ChatGPT Work. The discussion sets the stage for exploring recent updates and features, including the new voice interface, Heartbeats automation, ChatGPT Sites deployment, and his use cases for AI in content creation and video editing.

New Features in Codex and ChatGPT

Nick details recent enhancements in Codex and ChatGPT, notably the launch of the voice interface with a screen-reading orb that allows users to interact via voice commands seamlessly. He explains how Codex can manage multiple threads, fork new tasks, and handle complex workflows that were previously less accessible, emphasizing its potential to simplify and expand AI usage for both developers and general users.

Exploring ChatGPT Work and Heartbeats Automation System

The conversation covers the ChatGPT Work platform and its Heartbeats system, which automates routine checks like monitoring emails or financial transactions. Nick demonstrates how these systems enable continuous, hands-free oversight, with tasks running in the background and providing notifications or summaries, making AI-powered management practical and accessible via mobile and web.

Live Demonstration of Voice and Task Delegation

Nick showcases a real-time demo of using voice commands to manage complex tasks such as booking flights, hotels, and handling expense reports. This section illustrates how users can delegate multi-step workflows to AI with natural language, reducing manual effort and increasing productivity, emphasizing the importance of low latency and high understanding in voice interfaces.

Mobile AI Workflows and Content Creation with AI

The episode explores how ChatGPT on mobile transforms workflows, enabling continuous content curation, editing, and website deployment from a smartphone. Nick shares how he uses AI to gather, summarize, and publish content effortlessly, including deploying websites via ChatGPT Sites. He also discusses the advantages of AI-assisted video editing, where AI handles cuts, transcripts, and effects, easing the creation process for content creators.

AI-Assisted Video Editing and Content Management

Nick describes his use of plugins and AI tools to edit videos, generate highlights, blur sensitive information, and compile content efficiently. He emphasizes how these tools help solo creators by reducing the tedium of editing and allowing focus on creative aspects. The conversation highlights how AI automates the tedious parts of media production—transcription, selection of best takes, and assembling short clips.

Addressing Practical Concerns and Future Directions

The discussion ends with practical considerations, such as how voice interactions work around humans in office environments, and the potential future integrations like voice-enabled direct brain interfaces. Nick reflects on recent leaps in AI capabilities—especially those accessible via interfaces like ChatGPT Work and mobile apps—and predicts a growing emphasis on reducing latency to enhance real-time interactions.

Show notes

Nick Baumann is on the Developer Experience team at OpenAI, where he spends his days building with, testing, and communicating the capabilities of ChatGPT Codex and ChatGPT Work. In this episode, Nick walks me through several features that have launched or evolved recently: the new voice interface with its screen-reading orb, the Heartbeats automation system in ChatGPT Work on mobile, the live ChatGPT Sites deployment feature, and his personal use case for AI-assisted UGC video editing.

What you’ll learn:

How two-person voice chat works
How Heartbeats work
How to build and deploy a live website with ChatGPT Sites
How to delegate a flight search, hotel booking, and expense report to Codex in a single voice conversation without opening a single app manually
Why ChatGPT Work on mobile is the most underutilized AI workflow for people already using the ChatGPT app
How to use a custom UGC Video plugin to feed 50 raw clips into ChatGPT, let it pull transcripts, pick the best takes, and assemble a finished vertical video overnight

—

Brought to you by:

Bolt.new—Turn your idea into a real product

Hyperagent—Deploy fleets of agents that handle real work

—

In this episode, we cover:

(00:00) Introduction to Nick Baumann

(02:56) What’s new in Codex

(05:40) ChatGPT Work and Heartbeats

(06:40) Live Codex voice demo

(13:25) Latency vs. intelligence

(14:36) Quick recap