Comment on OpenAI Suspends ByteDance's Account After It Used GPT To Train Its Own AI Model

cmnybo@discuss.tchncs.de ⁨1⁩ ⁨year⁩ ago

• FaceDeer@kbin.social ⁨1⁩ ⁨year⁩ ago

  Ever since that paper about "model decay" this has been a common talking point and it's greatly misunderstood. Yes, if you just repeatedly cycle content through AI training over and over through successive generations, you get AIs that lose "fidelity." But that's not what any actual real world training regimen using synthetic data does. The helper AI is usually used to process input data. For example, if you're training an AI to respond in a chat-like format, you could take raw non-conversational text (like a book) and have the helper AI convert it into a conversation for the new AI to learn from. Or to take a real-world example, Dalle3 was trained by having a helper AI look at pictures and create detailed text descriptions of them to use as the caption to associate with the image when training.

  OpenAI has put these restrictions in its TOS as a way of trying to "pull up the ladder behind them", preventing rivals from trying to build AIs as good as the ones they have already. Fortunately it's not going to work. There are already open LLMs that can be used as "helpers" without needing OpenAI at all. ByteDance was likely just being lazy here.

  source
• ripe_banana@lemmy.world ⁨1⁩ ⁨year⁩ ago

  There is actually a whole subsection of AI focused on training one model with the output of another called knowledge distillation.

  source
• CaptainSpaceman@lemmy.world ⁨1⁩ ⁨year⁩ ago

  Like photocopying a picture of a terd

  source
• SeaJ@lemm.ee ⁨1⁩ ⁨year⁩ ago

  I’ve watched Multiplicity enough tones to know you get a slightly less functional copy.

  source

  • cybersandwich@lemmy.world ⁨1⁩ ⁨year⁩ ago

    She touched my peppy Steve.

    source
  • FaceDeer@kbin.social ⁨1⁩ ⁨year⁩ ago

    That wasn't a documentary, and it wasn't about machine learning.

    source
• altima_neo@lemmy.zip ⁨1⁩ ⁨year⁩ ago

  Works kinda neat with stable Diffusion tho

  source
• ech@lemm.ee ⁨1⁩ ⁨year⁩ ago

  Depends how it’s done. GAN (Generative Adversarial Network) training works with exactly that, having networks train against each other, each improving the other over time.

  source
• betterdeadthanreddit@lemmy.world ⁨1⁩ ⁨year⁩ ago

  Sounds like what you’d get if you ordered a ChatGPT off of Wish dot com. Cheap knock-offs that blatantly steal ideas/designs and somewhat work are kinda their thing.

  source
• ZickZack@fedia.io ⁨1⁩ ⁨year⁩ ago

  Not necessarily: there have been recent works that indicate that filtering effects of fine tuned LLMs greatly improves the data efficiency (e.g phi-1).
  Further, if you have e.g. human selection on top of LLM generated content you can get great results as the LLM generation can be used as a soft curriculum, with the human selection biasing towards higher quality.

  source