Comment by athyuttamre
9 hours ago
(Atty from OpenAI here)
GPT-Live-1 is the first version of a new generation of models, and we believe the full-duplex architecture + delegation enables entirely new ways of human-AI interaction.
Would love to hear your feedback!
Hey! Bit of an unusual question maybe: if this stuff further exarcerbates the loneliness epidemic and atomization of society, will you be able to live with yourself you think? If you hear about teenagers only spending time with your chatbot in 5 years, will you feel some amount of personal responsibility or not? Always curious to hear you guys' perspective on that kind of stuff!
> if this stuff further exarcerbates the loneliness epidemic and atomization of society, will you be able to live with yourself you think?
Looking at the 30,000-foot view of how society is set up: laws, economic system, employee incentives, etc, do you suppose it matters what the individual contributors think? I say this not to absolve anyone of responsibility, but to point out the obvious outcomes of our incentives across the strata (polity -> shareholders -> boards -> C-suite -> employees)
I will bet you dollars to donuts, somewhere inside OpenAI is a frequently-used revenue dashboard, but not for loneliness - if anything, OpenAI will make horny models and tout itself as a solution to loneliness, a la character.ai - if that earns them more money.
That's not really a fair question, especially given the audience here on HN. Most here have blood on their hands from the previous 25 years of tech-eats-the-world.
Technology is what you make of it. If you want to make it your best friend and your only friend, that's your choice. But I would guess more people would use it as a personal tutor. If it works as well as the demo, this already crushes most language learning apps. Actually the only apps for learning language that it would fail to replace would still be Tandem and HelloTalk because it still can't replicate the nuance of real human interaction.
"Blame the individual" doesn't really make sense when we're talking about things that have a society-wide impact. As an analogue, personally I'm not on social media, but I am affected by social media because everyone around me is on social media and their happiness and wellbeing also impacts me. AI companies need to stop pretending they can just outsource responsibility for their products.
I think assuming people will use it as a tutor/learning tool is.. way too optimistic. A small fraction will, but the majority will just view things like a second language as something not worth learning.
2 replies →
I would say no if the question was asked of me. It’s like asking if you’d feel bad about marriage if you ended up divorced with a child. Or if you’d feel bad about commenting if it made someone feel like they lost brain cells.
The morality of an action isn’t based on actual consequences because the future isn’t known in advanced. All we can do is act on the perceived consequences of our actions, and if we think those are good, pursue them.
You don't go into a marriage assuming you'll get divorced. You assume you won't, hope for the best, work not to, and then it happens. You work through those problems with the child so that it hopefully doesn't impact them as much as it otherwise could.
The loneliness epidemic is driven by companies maximizing keeping their customers engaged with their screens, something OpenAI is wont to do. Knowing that the company wants customers engaged and that this will do that, and also knowing that that plays into the loneliness epidemic by substituting human interaction, makes it far different than getting married and then maybe or maybe not getting divorced.
People who do this kind of stuff are very irritating. You clearly have some problem with the work they do. Instead of saying and approaching that outright, you pass it in some passive aggressive fake bullshit. Makes you sound like the kind of person I would much rather not be speaking to, which is kind of ironic given your comment.
i mean you chose to be irritated by it ¯\_(ツ)_/¯
"Hey Leibniz, how do you live with yourself knowing that your binary system helped eventually replace human conversations?"
Fair question, although I think he really has a hard time living with it ...
Oh no, now I'd better blame Hitler and Stalin's ancestors for their misdeeds. Of course, the SS soldier bears no responsibility, however, as he was just doing his job.
So many wet blankets on this post today with these types of questions...
Its weird how years of relentless doom trolling and threatening peoples livelihoods, mixed with absolutely real, measurable, AWFUL societal impacts have soured people!
I love the chatty tone of this utterly dystopian question.
I don't think the question is really the dystopian part
Can I connect it to my skills/tools? Example case, I have a knowledge base and event log in my company. I need a brainstorm companion, which will have full access to this knowledge, can converse about it and can invoke skills/tools available in the repo.
In ChatGPT, Voice doesn't yet support connectors, but we're hoping to add support soon! Once GPT-Live launches in the API, you can also build custom integrations yourself.
+1 to the op. This is one of the first things I thought about after using it for a few minutes. I have a large collection of local knowledge bases that I use through codex, having this real time thing be able to plug into it would be the killer feature, otherwise the utility is severely limited. I don't want to talk to it about the weather, I want to brainstorm about the things I am working on, which it currently knows nothing about.
Do integrations supporting streaming input?
One big gap I've run into for UX is most realtime voice harnesses wait for a full response from tools, and at most support the model filling the dead air until then
It'd be a game-changer to be able to have the model start replying with partial information streamed from the tool call, then seamlessly continue with additional information.
1 reply →
If it's not able to connect to my other apps like gcal, it should suggest that I use the standard gpt client, instead of telling me 'I don't have the capability to do that' which is misleading as some might presume the normal client doesn't have that capability either.
Can it delegate to just one agent at a time or can it spawn multiple subagents for different tasks?
There are many delegation models possible:
1. The voice model delegates to one agent.
2. The voice model delegates to multiple agents, and keeps track of tasks.
3. The voice model delegates to an orchestrator agent, which then delegates to sub-agents and keeps track of tasks.
YMMV depending on the exact product experience you care about, because there is a tradeoff between latency and layers of delegation.
Our current implementation is backed by one model, but you can imagine this getting much better with time.
Can it sing? Is it an end to end multimodal model?
I'm interested in how you can present simultaneous rich visual information about what is happening the side delegation work.
i.e. how will full duplex & delegation enable/enhance desktop flows w/o corresponding leaps in UI.
What made you to try again?
Can we have less terrible voices please? Nothing that sounds like a bubbly millennial. Literally anything that has gravitas.
Have you tried some of our deeper voices like Spruce? Would love to hear what your ideal voice is.
Does video/image input still work with these duplex models?
Image input is supported, but video is not today. We're working hard to bring it to you soon.
How does it compare to the realtime-2 model?
> GPT-Live-1 is the first version of a new generation of models, and we believe the full-duplex architecture + delegation enables entirely new ways of human-AI interaction.
Awesome. Are you guys able to share anything about the model architecture? I've been interested lately in split-transformer RVQ-based conversational agents, e.g. via stuff like https://arxiv.org/abs/2412.10208 (ResGen) and https://arxiv.org/abs/2603.18090 (MOSS-ITT) and of course Moshi (https://arxiv.org/pdf/2410.00037).
Intuitively, decoupling semantic and audio-timeslice-space generations with coupled but distinct histories is right model architecture, not just for these sorts of assistants, but for domains like robotics too.
When is it rolling out? Currently on ChatGPT Pro but not seeing it yet?
Are you still using LiveKit for the back-and-forth architecture
Write up about the architecture is here https://openai.com/index/delivering-low-latency-voice-ai-at-...
> Would love to hear your feedback!
I'm currently on the 20 $/mo subscription and using codex meaningfully, and i'm loving this.
I am considering bumping my subscription to the 100 $/month and this might be the reason i switch, BUT: i really envision me using this also through other means as well (eg: agents like openclaw/hermes) in agentic ways.
Will this be supported?
I can make OpenAI stuff the center of my agentic AI life, but I need it to be interoperable.
We're adding support to the API soon, which will let you integrate with any agent in the background. Would love to see the community go wild with it. You can sign up to be notified here: https://openai.com/form/gpt-live-1-in-the-api/
[dead]
- As models get better, have you considered some kind of filter or particular cadence to serve as a reminder that the user is not talking to a human?
- The videos felt scripted and dishonest