3 ms·

I'm confused by the term "whistleblower" here. Was anything actual released that wasn't publicly known? It seems like he just disagreed with whether it was "f

I'm confused by the term "whistleblower" here. Was anything actual released that wasn't publicly known? It seems like he just disagreed with whether it was "fair use" or not, and it was notable because he was at the company. But the facts were always known, OpenAI was training on public copyrighted text data. You could call him an objector, or internal critic or something.

stonogo 2y ago
The article holds clues: "Information he held was expected to play a key part in lawsuits against the San Francisco-based company."
- abeppu 2y ago
  and later: >In a Nov. 18 letter filed in federal court, attorneys for The New York Times named Balaji as someone who had “unique and relevant documents” that would support their case against OpenAI. He was among at least 12 people — many of them past or present OpenAI employees — the newspaper had named in court filings as having material helpful to their case, ahead of depositions. Yes it's true it's been public knowledge that OpenAI has trained on copyrighted data, but details about what was included in training data (albeit dated ...), as well as internal metrics (e.g. do they know how often their models regurgitate paragraphs from a training document?) would be important.
  - janalsncm 2y ago
    I guess the question is whether those documents have already been entered into evidence?
neuroelectron 2y ago
The issue is it has to be proven in court. This man was personally responsible for developing web scraping; stealing data from likely copyrighted sources. He would have had communications specifically addressing the legality of his responsibilities, which he was openly questioning his superiors about.
- ALittleLight 2y ago
  "Stealing data" seems pretty strong. Web scraping is legal. If you put text on the public Internet other people can read it or do statistical processing on it. What do you mean he was "stealing data"? Was he hacking into somewhere?
  - canoebuilder 2y ago
    In a lot of ways, the statistical processing is a novel form of information retrieval. So the issue is somewhat like if 20 years ago Google was indexing the web, then decided to just rehost all the indexed content on their own servers and monetize the views instead linking to the original source of the content.
    - sashank_1509 2y ago
      It’s not anything like rehosting though. Assume I read a bunch of web articles, synthesize that knowledge and then answer a bunch of question on the web. I am performing some form of information retrieval. Do I need to pay the folks who wrote those articles even though they provided it for free on the web? It seems like the only difference between me and ChatGPT is the scale at which ChatGPT operates. ChatGPT can memorize a very large chunk of the web and keep answering millions of questions while I can memorize a small piece of the web and only answer a few questions. And maybe due to that, it requires new rules, new laws and new definitions for the better of society. But it’s nowhere near as clear cut as the Google example you provide.
      - underbiding 2y ago
        I love this argument. "Seems like only difference between me and ChatGPT is absolutely everything". You can't be flippant about scale not being a factor here. It absolutely is a factor. Pretending that ChatGPT is like a person synthesizing knowledge is an absurd legal argument, it is absolutely nothing like a person, its a machine at the end of the day. Scale absolutely matters in debates like this.

stonogo 2y ago

The article holds clues: "Information he held was expected to play a key part in lawsuits against the San Francisco-based company."

abeppu 2y ago
and later: >In a Nov. 18 letter filed in federal court, attorneys for The New York Times named Balaji as someone who had “unique and relevant documents” that would support their case against OpenAI. He was among at least 12 people — many of them past or present OpenAI employees — the newspaper had named in court filings as having material helpful to their case, ahead of depositions. Yes it's true it's been public knowledge that OpenAI has trained on copyrighted data, but details about what was included in training data (albeit dated ...), as well as internal metrics (e.g. do they know how often their models regurgitate paragraphs from a training document?) would be important.
- janalsncm 2y ago
  I guess the question is whether those documents have already been entered into evidence?

neuroelectron 2y ago

The issue is it has to be proven in court. This man was personally responsible for developing web scraping; stealing data from likely copyrighted sources. He would have had communications specifically addressing the legality of his responsibilities, which he was openly questioning his superiors about.

ALittleLight 2y ago
"Stealing data" seems pretty strong. Web scraping is legal. If you put text on the public Internet other people can read it or do statistical processing on it. What do you mean he was "stealing data"? Was he hacking into somewhere?
- canoebuilder 2y ago
  In a lot of ways, the statistical processing is a novel form of information retrieval. So the issue is somewhat like if 20 years ago Google was indexing the web, then decided to just rehost all the indexed content on their own servers and monetize the views instead linking to the original source of the content.
  - sashank_1509 2y ago
    It’s not anything like rehosting though. Assume I read a bunch of web articles, synthesize that knowledge and then answer a bunch of question on the web. I am performing some form of information retrieval. Do I need to pay the folks who wrote those articles even though they provided it for free on the web? It seems like the only difference between me and ChatGPT is the scale at which ChatGPT operates. ChatGPT can memorize a very large chunk of the web and keep answering millions of questions while I can memorize a small piece of the web and only answer a few questions. And maybe due to that, it requires new rules, new laws and new definitions for the better of society. But it’s nowhere near as clear cut as the Google example you provide.
    - underbiding 2y ago
      I love this argument. "Seems like only difference between me and ChatGPT is absolutely everything". You can't be flippant about scale not being a factor here. It absolutely is a factor. Pretending that ChatGPT is like a person synthesizing knowledge is an absurd legal argument, it is absolutely nothing like a person, its a machine at the end of the day. Scale absolutely matters in debates like this.