Hacker News with Generative AI: Training Data

Ask HN: Is politeness towards LLMs good training data, or just expensive noise? (ycombinator.com)
Sam Altman recently said user politeness towards ChatGPT costs OpenAI "tens of millions" but is "money well spent."
Artificial Intelligence, Training Data, User Interface
8 points by scottfalconer 2 days ago | 12 comments
The Unbelievable Scale of AI's Pirated-Books Problem (theatlantic.com)
Meta pirated millions of books to train its AI. Search through them here.
Artificial Intelligence, Piracy, Books, Training Data
8 points by the-mitr 35 days ago | 0 comments
The Scale of AI's Pirated-Books Problem (theatlantic.com)
Meta pirated millions of books to train its AI. Search through them here.
Artificial Intelligence, Piracy, Books, Training Data
10 points by hakaneskici 37 days ago | 1 comments
There's No Longer Any Doubt That Hollywood Writing Is Powering AI (theatlantic.com)
Dialogue from these movies and TV shows has been used by companies such as Apple and Anthropic to train AI systems.
Artificial Intelligence, Hollywood, Movies, TV Shows, Training Data
11 points by doener 141 days ago | 6 comments
SwiGLU activation function causes instability in FP8 LLM training (arxiv.org)
We train, for the first time, large language models using FP8 precision on datasets up to 2 trillion tokens -- a 20-fold increase over previous limits.
Artificial Intelligence, Machine Learning, Training Data, Hardware
10 points by LarsDu88 217 days ago | 2 comments
Leaked Docs Show Nvidia Scraping a Human Lifetime of Videos per Day to Train AI (404media.co)
Artificial Intelligence, Data Scraping, Privacy, Nvidia, Training Data
50 points by depingus 264 days ago | 11 comments
Apple, Nvidia, Anthropic Used Swiped YouTube Videos to Train AI (proofnews.org)
Artificial Intelligence, Training Data, YouTube, AI Ethics, Copyright
41 points by gwintrob 283 days ago | 66 comments
YouTube creators surprised to find Apple and others trained AI on their videos (arstechnica.com)
YouTube, Artificial Intelligence, Copyright, Training Data, Content Creation
5 points by ourmandave 283 days ago | 3 comments
Figma will use your content to train its AI (stackdiary.com)
Figma, Artificial Intelligence, Training Data
20 points by tgol 300 days ago | 4 comments
OpenAI destroyed a trove of books used to train AI models (businessinsider.com)
OpenAI, Artificial Intelligence, Data Sets, Training Data
34 points by nitramm 352 days ago | 35 comments