4 ms·

And a notice that chats will be used for training.

seanhunter 3y ago
My understanding is they want to give themselves the option to do this in future even if they aren't doing it right now.
- slowmotiony 3y ago
  My understanding is none of us here has any understanding what they do or don't do. We literally have no idea what's going on inside OpenAI.
  - uLogMicheal 3y ago
    Such an open company... /s Edit RE below comment: The company was literally named that as it was started to be the "open" AI company, given the dangers of centralization of such tech. From Wikipedia -- The organization stated it would "freely collaborate" with other institutions and researchers by making its patents and research open to the public.[
    - seanhunter 3y ago
      Not sure why everyone keeps going on about the word "open" in the name openAI. Microsoft isn't small. Apple isn't a fruit. Berkshire Hathaway isn't from Berkshire and is nothing to do with Shakespeare's wife or the actress. Oracle isn't an ancient Greek priestess making enigmatic utterances from inside a temple. Blackrock isn't black or a rock. Citadel isn't any kind of fortress at all. Etc etc. It's just a name.
    - seanhunter 3y ago
      OpenAI publishes a huge amount of research and makes huge amounts of data available for research. They do collaborate with other institutions and researchers. eg: GPT-2: paper https://d4mucfpksywv.cloudfront.net/better-language-models/language-models.pdf https://d4mucfpksywv.cloudfront.net/better-language-models/l... with code: https://github.com/openai/gpt-2 https://github.com/openai/gpt-2, output dataset https://github.com/openai/gpt-2-output-dataset https://github.com/openai/gpt-2-output-dataset and a bunch of blog posts with all manner of details GPT-3: paper https://arxiv.org/abs/2005.14165 https://arxiv.org/abs/2005.14165 dataset https://github.com/openai/gpt-3 https://github.com/openai/gpt-3 etc etc.
      - uLogMicheal 3y ago
        but as soon as they mined gold they closed the information hose?
- throwuwu 3y ago
  My understanding is that Microsoft research has already published a paper where they used synthetic chat interactions of the same form that chatGPT uses to train a new model. GPT4 could be used to select the best interactions from which to create a training set. I’d be very, very surprised if OpenAI hasn’t already been doing this internally. https://arxiv.org/pdf/2306.02707.pdf https://arxiv.org/pdf/2306.02707.pdf

My understanding is they want to give themselves the option to do this in future even if they aren't doing it right now.

slowmotiony 3y ago
My understanding is none of us here has any understanding what they do or don't do. We literally have no idea what's going on inside OpenAI.
- uLogMicheal 3y ago
  Such an open company... /s Edit RE below comment: The company was literally named that as it was started to be the "open" AI company, given the dangers of centralization of such tech. From Wikipedia -- The organization stated it would "freely collaborate" with other institutions and researchers by making its patents and research open to the public.[
  - seanhunter 3y ago
    Not sure why everyone keeps going on about the word "open" in the name openAI. Microsoft isn't small. Apple isn't a fruit. Berkshire Hathaway isn't from Berkshire and is nothing to do with Shakespeare's wife or the actress. Oracle isn't an ancient Greek priestess making enigmatic utterances from inside a temple. Blackrock isn't black or a rock. Citadel isn't any kind of fortress at all. Etc etc. It's just a name.
  - seanhunter 3y ago
    OpenAI publishes a huge amount of research and makes huge amounts of data available for research. They do collaborate with other institutions and researchers. eg: GPT-2: paper https://d4mucfpksywv.cloudfront.net/better-language-models/language-models.pdf https://d4mucfpksywv.cloudfront.net/better-language-models/l... with code: https://github.com/openai/gpt-2 https://github.com/openai/gpt-2, output dataset https://github.com/openai/gpt-2-output-dataset https://github.com/openai/gpt-2-output-dataset and a bunch of blog posts with all manner of details GPT-3: paper https://arxiv.org/abs/2005.14165 https://arxiv.org/abs/2005.14165 dataset https://github.com/openai/gpt-3 https://github.com/openai/gpt-3 etc etc.
    - uLogMicheal 3y ago
      but as soon as they mined gold they closed the information hose?
throwuwu 3y ago
My understanding is that Microsoft research has already published a paper where they used synthetic chat interactions of the same form that chatGPT uses to train a new model. GPT4 could be used to select the best interactions from which to create a training set. I’d be very, very surprised if OpenAI hasn’t already been doing this internally. https://arxiv.org/pdf/2306.02707.pdf https://arxiv.org/pdf/2306.02707.pdf