I’m as curious and excited as any gadget lover to see what newfangled AI thingamabob will come from OpenAI’s $6.5 billion purchase of the ex-Apple designer Jony Ive’s IO company.
But I have one request: Please, for the love of God, do not make it a voice-controlled device.
Here’s what we know about the possible device that Ive and Sam Altman are teasing in a video about their new collaboration. The Wall Street Journal reported that Altman told OpenAI employees a few details:
The product will be capable of being fully aware of a user’s surroundings and life, will be unobtrusive, able to rest in one’s pocket or on one’s desk, and will be a third core device a person would put on a desk after a MacBook Pro and an iPhone.The Journal earlier reported that the device won’t be a phone, and that Ive and Altman’s intent is to help wean users from screens. Altman said that the device isn’t a pair of glasses, and that Ive had been skeptical about building something to wear on the body.
Ming-Chi Kuo, a supply chain analyst who is often correct about coming hardware, says that the device may be something larger than the Humane AI Pin, and possibly worn around the neck.
I am extremely nervous that this sounds like it might be some sort of voice-controlled device.
Don’t get me wrong: I’m an enthusiastic voice user of Alexa (at home) and Siri (in the car). I can see how convenient it is.
But the idea of talking to Siri while walking down the street or at a grocery store gives me hives. The idea of saying, “Hey Meta, take a picture” to activate my Ray-Bans while at a Benson Boone concert makes me want to bite my cyanide capsule. If I ever start using Siri out loud at my desk in the office, I fully accept that HR can fire me on the spot.
(Of course, voice-controlled devices are an accessibility issue for some people who are blind, have low vision, or otherwise have trouble using a screen device. I am not talking about this use, which is obviously good and a benefit. Perhaps society would be better if public use of voice devices were more normalized!)
OpenAI does seem to be interested in voice. At a demonstration over a year ago, it showed new voices that could talk to you (this was the demonstration that infamously got them in trouble with Scarlett Johansson for making a voice option suspiciously close to her own).
Meta has also embraced the idea of voice controls. Its stand-alone Meta AI app is meant for natural conversations between you and the app on your phone (though, at the moment, it’s laggy and often leads to crosstalk).
It’s a long-held sci-fi dream to have a supersmart AI agent you can just talk to naturally. Like Tony Stark’s Jarvis, or the ScarJo voice in “Her.” But even “Knight Rider” had the basic understanding that it was OK to talk to your car only in your car.
Humane’s AI Pin turned out to be a disaster, partly because it just didn’t work very well. Let’s assume whatever OpenAI/IO is cooking up will be good at doing what it’s supposed to.
Based on what the AI Pin could do, and other examples of AI assistants or devices, I can make a few guesses of what it might be able to do: listen to your work meeting and take notes for you, give you information about something you see in front of you (“What building is this?” or, “Which of these two sandwiches at Pret has more protein?”), do personal assistant tasks for you (“How much time until my next meeting?” or, “Text Hayley and tell her I’m running late.”)
These all sound great and convenient! But a lot of them require something that I don’t think I want to be doing: talking to my device out loud, in public, constantly.
We’re at a moment in society where people are already pushing the norms of what is appropriate public device use. People are watching TikTok without headphones on the subway. Texting at the movie theater. Filming themselves at the gym. It was already weird enough when people started talking on the phone with AirPods in, making it unclear whether the person headed toward you on the sidewalk was on a call or experiencing a religious revelation. I don’t think we’re ready for a world where people are constantly talking to their always-on, always-listening AI devices.
So I eagerly await this device, which Altman says will sell 100 million units and be ready by the end of 2026. But please, please, do not make me talk to it in public.
