As AI scales, progress is often measured in parameters, training data size, benchmark scores and datacenter construction. But how we interact with models matters too.

If you haven't started communicating with ChatGPT and other AI models simply by speaking to it, you should. It might change how you prompt in general.

Being “voicepilled” is that moment of realization that once you start seriously using your voice to interact with technology, you may never go back to only using keyboards ever again. If you use products like @WisprFlow or @ChatGPTapp voice, you know what I mean.

Human-computer interaction has always been a blend of art, psychology, and physiology. When DARPA first sponsored Douglas Engelbart’s team to create the mouse, it was a reimagining of what computing could feel like.

Search engines did something similar, using a combination of technology and interface design that ended up changing the way we query the world around us. In short, they altered our mental (and physical) relationship to knowledge itself.

Voice user interfaces do the same. We can use them to send quicker texts to friends, to query AI models, and eventually to interface with a Waymo as it takes you to your destination.

Will people abandon keyboards completely? Not if they spend a lot of their time working with spreadsheets, writing complex documents, working with imaging and video-editing software, or love communicating with GIFs and emojis.

But for many everyday purposes, voice is simply faster, more natural, and more flexible than typing. And what’s changed now is that state-of-the-art AI models can genuinely process what we say.

This can change how we prompt models like ChatGPT. Typing often nudges us to think about how we shorten our thoughts to conserve effort. Voice is iterative by default. You stumble, you rephrase, you interrupt yourself. You can do unstructured brain dumps: “just take this

So voice input can be a powerful antidote to perfectionism. It teaches you that iteration is the workflow.

Voice might even reshape hardware and the architecture of work. Right now, sitting next to someone dictating all day feels unusual (unless you’re one of the stenophone diehards I know).

As voice becomes primary, we may design office spaces knowing that most of our time will be spent in dialogue with our workstations, versus typing on them.

Every major leap in computing has arrived not just from smarter machines, but from better interfaces. As AI models scale, one of the most profound changes will be the fact that everyone can now interact with them in the most human way possible.

That means billions of people, across cultures and literacy levels, will have a lower barrier to participation in the AI world. To be voicepilled is to glimpse this future.

@reidhoffman Does everyone enjoy talking that much? I was just thinking about this recently. Maybe it’s just me, but I’d much rather interact via typing than speaking. Even with background noise like a fan or music, or that my dogs think I’m talking to them 🤣, it just doesn’t seem as

@reidhoffman Even just dictating into GPT and letting it turn long-form talk into clean text is a massive time saver. Everyone’s chasing home runs with voice, but there's so much value in boring, everyday tasks. Voice is a new medium to do stuff.

@reidhoffman

@reidhoffman We’re trying to build the voice interface for commerce, looks something like this in its MVP stage

@reidhoffman taking and even full natural written language is in efficient compared to syntax and short hand. We’re already seeing the push back with “cookbooks” that amount to json or xml syntax. Absolutely nonsensical, sorry.

@reidhoffman I was going tell you to try the open source Happy Coder app to give instructions to Codex with your voice while away from your desk.. But I clicked on your profile because you looked old and realized you are a billionaire, so not actually in the market.

@reidhoffman voice + “manage context” modality would be killer

@reidhoffman I use voice a lot because you just need to spill your thoughts and have the model craft something out of it You literally get to think out loud and have it clean up behind you

@reidhoffman Sesame is scary good. If you haven't chatted with Miles or Maya lately, check them out. Couple this with a frontier or "good enough" model and you have a game changer for a lot of people.

@reidhoffman voice is cool but typing feels sharper and more intentional, no awkward pauses or rambling needed

@reidhoffman using the voice feature on Chat, Claude & using WisprFlow have made engaging with both tech and real people SO much better

@reidhoffman Voice interfaces feel like the natural evolution from LinkedIn's social graph to the conversational graph. Just as you democratized professional networking, voice will democratize AI access. The magic isn't just talking to computers—it's that barriers dissolve when interaction

@reidhoffman Fascinating perspective on the future of voice interfaces. Curious to hear more about your vision for natural human-AI dialogue.

@reidhoffman This is the way

@reidhoffman Oratory ftw yet again

@reidhoffman Some interesting techniques should have us in a jiffy. 👍🍿🍿🍿

@reidhoffman Not that interested in what fat traitors wishing for the death of my President think about computing

@reidhoffman Human-Agentic is the key. Real human agency combined with controlled agentic functions and usage-based-payments. That’s what I’m building with NewsAgentAI and StreamAgentAI for consumers to access news and streaming content and pay for what they want, when they want it.

@reidhoffman Yup voice is the future. We are building that now with @AidellyAI https://x.com/wilzerjb/status/...

@reidhoffman Open Space limited… until we resolve this it’s not gonna happen soon.

@reidhoffman Couldn’t agree more - voice + AI is the biggest productivity hack in the history of knowledge work. Add personal memory and brand/business memory into the mix, and the speed you can work and move at is absurd.

@reidhoffman i think its a right time to build/explore personal ai assistance primarily accessible via voice.. thats what we are doing at @marfa_ai https://getmarfa.com