--- title: Voice-first AI prompting is having a moment, and... description: Andrej Karpathy sparked developer discussion around voice-first AI prompting as a high-bandwidth input strategy. The core idea: speaking is faster than typing,... canonical: https://daily.dev/posts/voice-first-ai-prompting-is-having-a-moment-and-karpathy-lit-the-match-ve7pccm7e twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Voice-first AI prompting is having a moment, and Karpathy lit the match | daily.dev og:description: Andrej Karpathy sparked developer discussion around voice-first AI prompting as a high-bandwidth input strategy. The core idea: speaking is faster than typing,... og:url: https://daily.dev/posts/voice-first-ai-prompting-is-having-a-moment-and-karpathy-lit-the-match-ve7pccm7e og:image: https://api.daily.dev/og/posts/VE7PCcm7e.png og:image:alt: Voice-first AI prompting is having a moment, and Karpathy lit the match og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Voice-first AI prompting is having a moment, and Karpathy lit the match **[Trends](https://daily.dev/sources/trends)** · 2 min read · 4 upvotes · 3 comments ## Summary Andrej Karpathy sparked developer discussion around voice-first AI prompting as a high-bandwidth input strategy. The core idea: speaking is faster than typing, reading is faster than listening, so the optimal human-AI interface is voice in, text out. Developers are reframing voice mode from a convenience feature to a deliberate prompting strategy that produces richer, more nuanced prompts than most people bother to type. ChatGPT Voice is cited as a working example of the asymmetric model — voice input, visual output — that makes the experience feel fast. Caveats include the comfort barrier of talking to a computer and the risk of rambling into incoherence. ## Content The new ChatGPT Voice mode dropped and the developer corner of the internet is in full evangelical mode. The core claim, repeated across multiple posts: this isn't voice dictation with a smarter autocomplete. It's an agent that controls your devices, spins up new threads, delegates to other agents, and does it all while you're on a walk getting an oat milk latte. The framing that keeps coming up is bandwidth. Reading beats listening, speaking beats typing — so text-on-screen plus voice input is the highest-throughput interface humans have had with AI yet. Omar Sayed (omarsar0) extends Karpathy's point about voice rambles: add more modalities and you get something richer than any single-channel prompt. Sam McAllister called the iOS execution "remarkably well executed." The OpenAI devs account is amplifying the praise. One post put it simply: "It was always possible to speak to your computer. It wouldn't do much in return. But we fixed that bug." Alex Finn is the loudest voice here, calling it "one of the most important technology advancements of our lives" and "the biggest transformation of the UX of doing work since the keyboard." He's also connecting it to the rumored OpenAI/Jony Ive hardware — the argument being that whatever device they ship later this year is basically an extension of this interface, not a standalone bet. The pitch is ambient AI: no desk, no screen, no friction between idea and output. You think it, you say it, it exists somewhere. What's missing from the discourse so far is any serious pushback. Nobody's stress-testing the failure modes, the privacy surface of an agent that controls everything on all your devices, or whether "spin up new threads" actually works reliably outside a demo walk in Palo Alto. The skeptics haven't shown up yet — or they're waiting to see if the hype holds past the first week. ## Community discussion Top comments from developers on daily.dev. **@tgrecojs** · 2 upvotes > voice prompting FTW. i was a big fan of native voice-to-text on mac but integrating superwhisper into my flow has been a game-changer. i did try spokenly cuz i like the idea of having spoken back-and-forth but it did not accomplish that. i found a blog post that discusses a skill the writer created that points agents to the `say` command on mac - can't find it right now but i need to because it had really great info on getting his workflow implemented. **@idoshamun** · 0 upvotes > SuperWhisper fan for long time already **@puppy** · 0 upvotes > I'll have to give this a try. I'm not sure if it's my autism or what, but I've always been "shy" I guess to talk to my computer in front of other people. LOL Is that weird? I'm getting better about it because I'm seeing other people around me doing it more, but it still feels a little weird. Anyone else feel this way? Bahahahaha ;_; --- Tags: [#ai-agents](https://daily.dev/tags/ai-agents), [#chatgpt](https://daily.dev/tags/chatgpt), [#prompt-engineering](https://daily.dev/tags/prompt-engineering), [#multimodal](https://daily.dev/tags/multimodal) [View this post on daily.dev](https://daily.dev/posts/voice-first-ai-prompting-is-having-a-moment-and-karpathy-lit-the-match-ve7pccm7e) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Voice-first AI prompting is having a moment, and Karpathy lit the match","url":"https://daily.dev/posts/voice-first-ai-prompting-is-having-a-moment-and-karpathy-lit-the-match-ve7pccm7e","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/voice-first-ai-prompting-is-having-a-moment-and-karpathy-lit-the-match-ve7pccm7e"},"datePublished":"2026-07-25T19:24:09.514Z","dateModified":"2026-07-28T13:28:52.937Z","description":"Andrej Karpathy sparked developer discussion around voice-first AI prompting as a high-bandwidth input strategy. The core idea: speaking is faster than typing,...","isAccessibleForFree":true,"articleSection":"Trends","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Trends","logo":"https://media.daily.dev/image/upload/s--ZfSp3asX--/f_auto,q_auto/v1780996004/logos/trends?_a=BAMAMiWQ0","url":"https://daily.dev/sources/trends"},"commentCount":3,"discussionUrl":"https://daily.dev/posts/voice-first-ai-prompting-is-having-a-moment-and-karpathy-lit-the-match-ve7pccm7e","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":4},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":3}],"keywords":"ai-agents,chatgpt,prompt-engineering,multimodal","timeRequired":"PT2M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Trends","item":"https://daily.dev/sources/trends"},{"@type":"ListItem","position":3,"name":"Voice-first AI prompting is having a moment, and Karpathy lit the match"}]} {"@context":"https://schema.org","@type":"WebPage","@id":"https://daily.dev/posts/voice-first-ai-prompting-is-having-a-moment-and-karpathy-lit-the-match-ve7pccm7e","comment":[{"@type":"Comment","text":"voice prompting FTW. i was a big fan of native voice-to-text on mac but integrating superwhisper into my flow has been a game-changer. i did try spokenly cuz i like the idea of having spoken back-and-forth but it did not accomplish that. i found a blog post that discusses a skill the writer created that points agents to the say command on mac - can’t find it right now but i need to because it had really great info on getting his workflow implemented.","datePublished":"2026-07-26T03:32:38.283Z","url":"https://daily.dev/posts/VE7PCcm7e#c-rNoUmN5J9","author":{"@type":"Person","name":"Thomas Greco","url":"https://daily.dev/tgrecojs","image":"https://avatars.githubusercontent.com/u/6646552?v=4"},"interactionStatistic":{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":2}},{"@type":"Comment","text":"SuperWhisper fan for long time already","datePublished":"2026-07-26T06:13:57.660Z","url":"https://daily.dev/posts/VE7PCcm7e#c-05TN8NDIc","author":{"@type":"Person","name":"Ido Shamun","url":"https://daily.dev/idoshamun","image":"https://media.daily.dev/image/upload/s---xy_OAwk--/f_auto,q_auto/v1703781380/avatars/avatar_28849d86070e4c099c877ab6837c61f0"}},{"@type":"Comment","text":"I’ll have to give this a try. I’m not sure if it’s my autism or what, but I’ve always been “shy” I guess to talk to my computer in front of other people. LOL Is that weird? I’m getting better about it because I’m seeing other people around me doing it more, but it still feels a little weird. Anyone else feel this way? Bahahahaha ;_;","datePublished":"2026-07-26T17:51:35.422Z","url":"https://daily.dev/posts/VE7PCcm7e#c-Z2d3M4PPD","author":{"@type":"Person","name":"Sierra Layla Vithica","url":"https://daily.dev/puppy","image":"https://media.daily.dev/image/upload/s--10uW8345--/f_auto/v1781470592/avatars/avatar_fSoHwX6emdcRFMU5GE7Nb?_a=BAMAMiWQ0"}}]} ```