2y ago

Valve Reveals Its Thought Process In Regard to Games Being Shipped With AI-Generated Content on Steam

www.ign.com Valve Reveals Its Thought Process In Regard to Games Being Shipped With AI-Generated Content on Steam - IGN

Following reports of games made with AI art being banned from Steam, Valve has revealed a bit of its thought process in regard to games being shipped with AI-generated content on its digital storefront.

You're viewing a single thread.

11 comments

ChatGPT summary of the article:
Valve, the digital storefront for games, has responded to reports of games made with AI-generated content being banned from Steam. In an official statement, Valve explained that it is still learning about AI technology and wants to ensure developers have the proper rights for all assets used in their games.
The company aims to ship as many titles as possible but faces challenges in verifying rights and legal uncertainty surrounding data used to train AI models. Valve clarified that its review process reflects current copyright laws and policies, not personal opinions. They encourage innovation and the use of AI technology within appropriate legal boundaries.
Valve also mentioned that while app-submission credits are usually non-refundable, they are willing to offer them in cases where the review process needs improvement.
- ChatGPT summary of the article:
  Ironic, isn't it.
- There should be a generated summary for every article that get posted, much appreciated :)
  
  Ooooo, a bot with a webhook for ChatGPT that summarizes articles. Reddit had something like that didn't they? Ya know, before they killed off 3rd party apps, and bots using the service.
  
  Yeah, I like the tldr bot.
  
  No!
  
  Would be better to not use chatgpt at all and use an open solution like open assistant.
  
  How does Open Assistant compare to something like GPT3.5? I ask because I recently use GPT-J, and it I can't imagine ever going back to it knowing how much better GPT3.5 is.
  
  Hard to say since I don't keep up with openAI's stuff anymore but from what I've seen it's pretty similar and a positive is that it censors itself less which results in more useful answers. You'd have to try it yourself to compare the newest versions and use cases. The website doesn't have internet access so you would have to deploy a version of open assistant on your machine to test it out fully.
  
  Okay, I might check it out.Though, I'm not so certain I'd be able to run that kind of thing on my rig, it's pretty beefy but I'm not sure if it's Neural Pathway kind of beefy haha

11 comments