ChatGPT for Android – Pre-orders
play.google.com
play.google.com
From time to time I've been chatting with hey pi [1] which you can add as a WhatsApp contact. It's pretty fun
Look, if you don't like the app or the website, you can ask the LLM to write you a new one.
This is OpenAI. Not Reddit.
> Is there any point in this beyond personal data spying?
Alternatively, this would limit the eyes on your inputs / outputs to just OpenAI instead of them and N random app vendors.
Altman was on the reddit board for a long time until relatively recently
and it would never be expected to cover its costs (let alone make a profit)
completely different to reddit
I get the cynicism with apps like threads, being mobile-app only.
But OpenAI's business model is selling API access to a LLM - chatGPT is just a basic consumer front end. They've given no indication that they would want to deprecate the very actively developed web front end - and I see no motivation for them to do that.
Unlike social media platforms - OpenAI doesn't need to control the consumer UI - they make money either way. Alternative UI's and adblockers aren't a threat to them. They also don't have "engagement stats" to worry about.
I'm just saying we shouldn't get pitch forks out just yet, "native app" is often a selling point.
Yeah, pretty much. The privacy controls on Android, at least, are very poor and you leak tons of private data to native apps that doesn't leak to web apps.
I personally am only against apps that give me zero utility or value over what a website can do. If I want to add you to my home screen, I will; if I care enough I'll certainly remember the website.
What does a native app bring to the table?
ChatGPT's engagement stats are the whole reason the latest LLM hype cycle began.
With a web app I can pretty much trust the browser to keep the cache at around 500MB for all pages I visit.
In fact, it’s easier to be tracked across the web than by individual apps without permission
> My concern with native apps is only sometimes about privacy
An app at least on iOS has no more access to information on your phone than a website unless you give it permission.
Not suddenly. I turned against mobile apps a long time ago, for privacy reasons. That said, I'm not upset about OpenAI releasing one. They can do whatever they like, obviously.
Of course, web sites or web apps aren't really any better when it comes to Android.
On a desktop, I prefer native applications over web sites or web apps because I can at least firewall off native applications.
What type of “telemetry” is available to an app connecting to remote servers without your permission that isn’t available to a website?
I mean I wouldn’t say they make money. They are losing millions and only surviving because of subsidies from Microsoft
Keeps you logged in, copy&paste is better, doesn't close the session with no way to continue the current completion when you lock your phone, etc.
Plus you can use Siri to Ask ChatGPT stuff. Can’t do this through the web interface.
https://help.openai.com/en/articles/7993358-chatgpt-ios-app-...
When I talk to ChatGPT in the app (using Whisper) it understands 100%. But Siri transcribes pretty much every single query wrong. ChatGPT shows superhuman level of understanding the messed up queries to even respond with anything coherent.
Shame that the app doesn't read the completions back to you. There are probably alternative frontends that do that, I will look for them. It would be very useful when driving/cooking/etc.
Sorry for the rant :D
I for one like installing and using native apps of services I use often. But if you prefer to do everything in the browser that's fine as well.
their customers expect an app. not having one is a risk.
Sharing data with OpenAI it's *irresponsible step* at this time of development.
any guides on how to do that?
You have two options:
1- Find some port for it to run locally on GitHub (it will be quantized and not that useful right now, till 2024-till Qualcomm will ship their phones). - see https://github.com/Bip-Rep/sherpa
*The better one* 2- Host the version on a server on the cloud (or locally on your computer with tunneling) using ooga-booga (with --api flag), and communicate with that self-hosted llama, directly. This way you won't be limited to 7/13B versions but can run the 70B...
No need to thank me...