Skip to main content
How I AI

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

41 min episode · 2 min read
·
Nick Baumann

Episode

41 min

Read time

2 min

Topics

Artificial Intelligence, Software Development, Science & Discovery

AI-Generated Summary

Key Takeaways

  • Voice-to-agent delegation: ChatGPT Voice (triggered via hotkey) can simultaneously manage 5-6 independent threads, each running separate tasks like booking flights, filing expense reports, and browsing the web — all in parallel. Unlike previous voice modes that used weaker models, this implementation runs full-intelligence models capable of autonomous multi-step execution without user re-engagement.
  • Thread orchestration as a hidden feature: ChatGPT can create new threads, message existing ones, and have threads communicate with each other — a capability most users never discover organically. Baumann recommends letting the model decide when to fork tasks into separate threads, as this preserves context quality and enables cleaner parallel processing across complex, multi-part workflows.
  • ChatGPT Sites for shareable artifacts: ChatGPT can now deploy live websites at chatgpt.site with SQL database support, S3 file storage, and environment variable configuration. Access control is granular — fully public, fully private, or restricted to specific email addresses with ChatGPT login. Baumann built a full content aggregation site using only mobile ChatGPT with its in-app browser.
  • UGC video editing via Codex plugin: Dumping 50-60 raw video clips into ChatGPT Work, combined with a custom plugin encoding format preferences (9x16, 4x5, caption rules, safety zones), produces edited vertical videos including automatic transcript extraction, take selection, blur/redaction of sensitive content, and caption placement — without manual review of individual clips.
  • Latency over intelligence for voice workflows: When AI has strong background tools to delegate to, response latency matters more than raw model intelligence in voice interactions. Baumann uses high-intelligence mode for planning tasks but prioritizes low-latency responses for real-time voice orchestration, arguing that users abandon functional AI workflows primarily due to spinners and delays, not output quality gaps.

What It Covers

Nick Baumann from OpenAI demonstrates advanced ChatGPT workflows including voice-activated multi-thread delegation, browser-based task automation, ChatGPT Sites deployment, and AI-powered video editing using Codex — showing how these combined primitives create an always-on assistant capable of managing complex parallel tasks hands-free.

Key Questions Answered

  • Voice-to-agent delegation: ChatGPT Voice (triggered via hotkey) can simultaneously manage 5-6 independent threads, each running separate tasks like booking flights, filing expense reports, and browsing the web — all in parallel. Unlike previous voice modes that used weaker models, this implementation runs full-intelligence models capable of autonomous multi-step execution without user re-engagement.
  • Thread orchestration as a hidden feature: ChatGPT can create new threads, message existing ones, and have threads communicate with each other — a capability most users never discover organically. Baumann recommends letting the model decide when to fork tasks into separate threads, as this preserves context quality and enables cleaner parallel processing across complex, multi-part workflows.
  • ChatGPT Sites for shareable artifacts: ChatGPT can now deploy live websites at chatgpt.site with SQL database support, S3 file storage, and environment variable configuration. Access control is granular — fully public, fully private, or restricted to specific email addresses with ChatGPT login. Baumann built a full content aggregation site using only mobile ChatGPT with its in-app browser.
  • UGC video editing via Codex plugin: Dumping 50-60 raw video clips into ChatGPT Work, combined with a custom plugin encoding format preferences (9x16, 4x5, caption rules, safety zones), produces edited vertical videos including automatic transcript extraction, take selection, blur/redaction of sensitive content, and caption placement — without manual review of individual clips.
  • Latency over intelligence for voice workflows: When AI has strong background tools to delegate to, response latency matters more than raw model intelligence in voice interactions. Baumann uses high-intelligence mode for planning tasks but prioritizes low-latency responses for real-time voice orchestration, arguing that users abandon functional AI workflows primarily due to spinners and delays, not output quality gaps.

Notable Moment

Baumann described holding his phone up to his laptop microphone so his mother could dictate an administrative problem directly to ChatGPT Voice — bypassing her own technical hesitation entirely. The model processed her unstructured speech and completed the task, illustrating that voice input doesn't require the user to be the one operating the device.

Know someone who'd find this useful?

Episode Transcript

If we go back to, you know, 2022, 2023, around the release of ChatGPT, I think a lot of people felt the AI then. A chatbot that I say things to and it intelligently says things back to me. I think with the advent of Cody agents and, like, the Cody agent harness in particular, now this AI is going out and, like, reading files and trading files and, like, running commands and it this is crazy. I sell one thing, and then it figures out all the other things. How does that sound? I haven't gone deep with voice. I wanna see it from the pro. How should we be using it? What's cool about it? So I'm just gonna trigger voice. I've got a hot key for it. I'm listening. Go ahead. K. There should be some Amazon receipt for a a mic I just bought recently. It's for DX uses. I'm hoping you can help me expense it. Could you spin up another thread to handle taking care of that expense report for me? Me? The expense task is running. It'll find the Amazon receipt and stage the Navon report, but not submit it until you confirm. I'll let you know when it's ready to review. If anybody has been lucky enough to have an assistant, this is how it gets done. And it is just, like, such a nice delegation experience to be like, hey. Can you take care of this? Hey. Can you fix this? Tell me what's going on. Typically, voice is great. You can, like, talk back and forth, but it's usually with a less strong model that is less capable. Whereas this, it's fully able to delegate and manage these fully, you know, five, six sole threads on its own, which is great because then you can have it do things on your behalf. Hot take question. What do you think matters more on voice experience latency or intelligence? Welcome back to How I AI. I'm Clarivo, product leader and AI obsessive here on a mission to help you build better with these new tools. Today, we have Nick Baumann at OpenAI, and he's gonna show us some of the advanced use cases of ChatChippity codecs and Chachipity work, including how you can talk to your computer to book your flights, use Chachipity sites to build websites you can share with anybody or nobody, and use my favorite workflow, which is editing creator content in codex. Let's get to it. This episode is brought to you by bolt. New, the AI app builder for people who have ideas and want to ship them. Most AI tools spit out code that looks great in a demo and falls apart the second you try to do anything real with it, or they lock you into their own platform with no real way out. Bolt is different. You describe what you wanna build, a startup MVP, a landing page, an internal tool, a side project, …

Get the full transcript (7,944 words) + summary by email — free

One-time email with the complete transcript and AI summary of this episode. No account needed.

One email, no spam. We’ll also show you what SignalCast does.

Browse all How I AI transcripts →

You just read a 3-minute summary of a 38-minute episode.

Get How I AI summarized like this every Monday — plus up to 2 more podcasts, free.

Pick Your Podcasts — Free

Keep Reading

Books, tools, and gear mentioned in this episode

SignalCast may earn commission on purchases via these links.

Tools

  • SPONSORS: HyperAgent (https://hyperagent.com/howiai)
  • ChatGPT VoiceRecommendedBy guest

    by OpenAI

    ChatGPT Voice (triggered via hotkey) can simultaneously manage 5-6 independent threads, each running separate tasks like booking flights, filing expense reports, and browsing the web — all in parallel.
  • ChatGPT SitesRecommendedBy guest

    by OpenAI

    ChatGPT can now deploy live websites at chatgpt.site with SQL database support, S3 file storage, and environment variable configuration.
  • CodexRecommendedBy guest

    by OpenAI

    UGC video editing via Codex plugin: Dumping 50-60 raw video clips into ChatGPT Work, combined with a custom plugin encoding format preferences (9x16, 4x5, caption rules, safety zones), produces edited vertical videos.
  • SPONSORS: Bolt (https://bolt.new/howiai)

More from How I AI

We summarize every new episode. Want them in your inbox?

Similar Episodes

Related episodes from other podcasts

Explore Related Topics

This podcast is featured in Best AI Podcasts (2026) — ranked and reviewed with AI summaries.

Read this week's AI & Machine Learning Podcast Insights — cross-podcast analysis updated weekly.

You're clearly into How I AI.

Every Monday, we deliver AI summaries of the latest episodes from How I AI and 192+ other podcasts. Free for one show.

Start My Monday Digest

No credit card · Unsubscribe anytime