{
  "id": 663656,
  "title": "Usage of Synthetic Data",
  "url": "/competitions/deep-past-initiative-machine-translation/discussion/663656",
  "author_name": "",
  "post_date": "2025-12-19T11:31:20.527830800Z",
  "votes": 9,
  "comment_count": 2,
  "views": 0,
  "content": "<p>Are we allowed to generate synthetic data for our trainings? If yes, does this allowance also include commercial services like Gemini and ChatGPT as generators &amp; editors?</p>",
  "messages": [
    {
      "id": "3379215",
      "postDate": "12/19/2025 11:31:20",
      "content": "<p>Are we allowed to generate synthetic data for our trainings? If yes, does this allowance also include commercial services like Gemini and ChatGPT as generators &amp; editors?</p>",
      "rawMarkdown": "Are we allowed to generate synthetic data for our trainings? If yes, does this allowance also include commercial services like Gemini and ChatGPT as generators & editors?",
      "votes": null
    },
    {
      "id": "3380515",
      "postDate": "12/22/2025 14:32:39",
      "content": "<p>From previous experiences, yes you can use them… while training phase you are not required to upload any script so no body know.</p>",
      "rawMarkdown": "From previous experiences, yes you can use them... while training phase you are not required to upload any script so no body know.",
      "votes": null
    },
    {
      "id": "3380517",
      "postDate": "12/22/2025 14:39:20",
      "content": "<p>Hi <a href=\"https://www.kaggle.com/nlztrk\" target=\"_blank\">@nlztrk</a>,</p>\n<p>Yes, this is generally fine, though pay attention to usage restrictions. (You would want to investigate whether Gemini, ChatGPT, etc. allow this use-case, for instance.) Prizewinners need to open source any data or code used during training as a condition to receiving prize money, as well, with some exceptions. See this section and the following in the rules for the details: <a href=\"https://www.kaggle.com/competitions/deep-past-initiative-machine-translation/rules#5.-winner-license\" target=\"_blank\">https://www.kaggle.com/competitions/deep-past-initiative-machine-translation/rules#5.-winner-license</a></p>",
      "rawMarkdown": "Hi @nlztrk,\n\nYes, this is generally fine, though pay attention to usage restrictions. (You would want to investigate whether Gemini, ChatGPT, etc. allow this use-case, for instance.) Prizewinners need to open source any data or code used during training as a condition to receiving prize money, as well, with some exceptions. See this section and the following in the rules for the details: https://www.kaggle.com/competitions/deep-past-initiative-machine-translation/rules#5.-winner-license",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 3380515,
      "author_name": "fyenneyenn",
      "author_url": "",
      "post_date": "12/22/2025 14:32:39",
      "content": "<p>From previous experiences, yes you can use them… while training phase you are not required to upload any script so no body know.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 3380517,
      "author_name": "ryanholbrook",
      "author_url": "",
      "post_date": "12/22/2025 14:39:20",
      "content": "<p>Hi <a href=\"https://www.kaggle.com/nlztrk\" target=\"_blank\">@nlztrk</a>,</p>\n<p>Yes, this is generally fine, though pay attention to usage restrictions. (You would want to investigate whether Gemini, ChatGPT, etc. allow this use-case, for instance.) Prizewinners need to open source any data or code used during training as a condition to receiving prize money, as well, with some exceptions. See this section and the following in the rules for the details: <a href=\"https://www.kaggle.com/competitions/deep-past-initiative-machine-translation/rules#5.-winner-license\" target=\"_blank\">https://www.kaggle.com/competitions/deep-past-initiative-machine-translation/rules#5.-winner-license</a></p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "3379215": "Are we allowed to generate synthetic data for our trainings? If yes, does this allowance also include commercial services like Gemini and ChatGPT as generators & editors?",
    "3380515": "From previous experiences, yes you can use them... while training phase you are not required to upload any script so no body know.",
    "3380517": "Hi @nlztrk,\n\nYes, this is generally fine, though pay attention to usage restrictions. (You would want to investigate whether Gemini, ChatGPT, etc. allow this use-case, for instance.) Prizewinners need to open source any data or code used during training as a condition to receiving prize money, as well, with some exceptions. See this section and the following in the rules for the details: https://www.kaggle.com/competitions/deep-past-initiative-machine-translation/rules#5.-winner-license"
  },
  "source": "meta"
}