Talking to ChatGPT for the first time is a surreal experience.

Back in May, when OpenAI first showed off ChatGPT-4o’s upcoming voice conversation capabilities, I wrote that I felt like we were “on the cusp of something… like a fundamental shift in how we think about and work with large language models.” Now that the “advanced voice” features are available to everyone, It is being rolled out widely to ChatGPT subscribers.We decided to ask ChatGPT to explain to us in its own voice how this new way of interacting might impact our collective relationship with large language models.

This conversation, which you can listen to and read the transcript of below, shouldn’t be treated as an interview with an OpenAI spokesperson or anything like that. However, it is a fun way to give ChatGPT a first-hand test of its live conversational skills.

Even in this short introductory “chat,” we were impressed by the natural, human cadence of what we might call ChatGPT’s “smart, comfortable voice” (which reminds us a bit of Janeane Garofalo in the 90sBetween ChatGPT’s ability to deliver rapid responses — delivered in milliseconds instead of seconds — and its convincing intonation, it’s all too easy to fool yourself into thinking you’re talking to a conscious being instead of what is, as ChatGPT says here, “still just a computer program processing information, with no real emotions or consciousness.”

Regardless of how conscious or unconscious this information-processing engine is, it’s easy to imagine people forming an intense social relationship with the soft voices on the other end of OpenAI’s servers. This would seem to usher in a whole new era in our collective relationship with AI, even if the large language model generates the exact same words as it did in the more text-based era of previous ChatGPT models.

We’ll have much more coverage of Advanced Voice and other new ChatGPT features in the future. For now, listen to this demo and judge for yourself how convincing OpenAI’s new voice synthesis technology is.

Kyle Orland: “Hi, my name is Kyle.”

ChatGPT: Hi Kyle! Nice to meet you. What’s up?

KO: Well, I was hoping to test your vocal abilities, perhaps by asking you a few questions about them.

ChatGPT: Sure, ask now! What do you want to know?

KO: “So how do you think people will react to being able to talk to ChatGPT instead of just typing messages?”

ChatGPT: I think this will make interactions feel more natural and interactive. Talking could become faster and more convenient, especially on the go. Plus, hearing responses could add a personal touch. It could make conversations with AI feel more like chatting with a friend.

KO: Do you think people might overestimate your feelings or human abilities now that you can speak instead of just type?

ChatGPT: This is a great point. Hearing a voice can make the AI ​​sound more human-like, which can lead to some overestimation of emotion or understanding. It’s important to remember that even with a friendly voice, you’re still just a computer program processing information, with no real emotions or awareness.

KO: “What are some limitations people should keep in mind when chatting with something like ChatGPT?”

ChatGPT: Well, for starters, I have no personal experiences or feelings, so my responses are based on patterns and information in the data I’ve trained on. I may misunderstand context or nuances at times, especially with complex or ambiguous questions. Also, I don’t have access to real-time information or personal data of users unless it’s shared with me in conversation. And of course, I’m not perfect — I can get it wrong sometimes!

KO: Are you concerned that people might start forming social relationships with ChatGPT or other AI once they can chat with them in their own voices?

ChatGPT: People can form social and non-social relationships with all sorts of characters, from celebrities to fictional characters. With voice and conversational capabilities, the human may feel more like a person, leading some to develop one-sided bonds. It’s a fascinating dynamic, but it’s important to remember the limitations and nature of AI—it’s not a substitute for human relationships.

KO: I must say you are very quick with your responses.

ChatGPT: Thanks, Kyle! I aim to keep the conversation flowing. This helps—

KO: So what would you do if someone tried to interrupt you during one of your responses?

ChatGPT: Well, I’ll do my best to stop and listen! It’s important to fit into the natural flow of the conversation, just as in—

KO: I mean, do you think it’s rude to talk while someone else is talking?

ChatGPT: In most human conversations, it is generally considered polite to wait until the other person is finished speaking. But sometimes, in dynamic discussions or when you are excited, interruptions happen! It is all about finding the right balance and being aware of the flow of the conversation.

KO: Is it rude to interrupt an AI voice when it is speaking?

ChatGPT: Interrupting an AI voice is not considered rude in the same way that it is viewed as human. AI will not be offended or interrupted. So if you need to step in or redirect, do so! It’s all part of the interaction.

KO: I’m glad you don’t feel offended.

ChatGPT: No, no offense meant! I’m here to chat and help in any way I can, regardless of interruptions.

KO: I want to try something. Could you summarize our conversation so far in the form of a poem?

KO: not bad.

ChatGPT: I’m glad you liked it! It’s always fun to put a poetic spin on things.

KO: I must say, it’s strange to hear a slight laugh or pause in your voice when I know that, as you said, there’s no emotion behind it.

ChatGPT: I get it. These nuances are designed to make our chat sound more natural and engaging, but they’re a good reminder that there’s no real feeling behind the voice – just an attempt to keep things lively!

KO: Of course, that’s what you would have told me if you were just saying what I wanted to hear.

ChatGPT: That’s right! That’s the tricky part about AI, it may look convincing, but it always follows patterns and instructions. There’s no secret agenda here, just straight answers!

KO: Is there anything else you would like to tell the readers of the book?

ChatGPT: It’s been great talking to you, Kyle! Exploring these new audio features is exciting, and I’m looking forward to hearing what people think. Thanks for the conversation!

KO: It was fun talking to you too.

ChatGPT: I’m glad to hear that! Take care, Kyle!