Gary Marcus

Rank 27 of 47
|
Score 8
6/28/2024, 10:56:00 PM
X
In reply to:
Ryan Hickman
@ryanmhickman
·
121d
@Rahll @EL_EH_OH @GaryMarcus @marcus @rahlll @IEEESpectrum You can watch a movie once and have a rough recollection of it forever.

Of course you can't put on a Batman costume and act out The Dark Night for profit, but half of Netflix is derivative crap made by humans that's only slightly different.

Generative AI is just like that.
Reid Southen
@Rahll
·
121d
@ryanmhickman @EL_EH_OH @GaryMarcus @marcus @rahlll @IEEESpectrum Also, buying a copy of a movie doesn't grant you a license to do what you want with it, come on. There's even strict rules on being able to publicly display or screen a movie you bought, and you think you're allowed to just rip it into a model? Get outta here.
Ryan Hickman
@ryanmhickman
·
121d
@EL_EH_OH @GaryMarcus @Rahll @marcus @rahlll @IEEESpectrum IMO, humans are welcome to learn from works they observe, and so are the AI systems humans are creating to supercharge the next generation of artists

In some cases companies may need to buy a copy of a movie to learn from it
Leo
@EL_EH_OH
·
121d
@ryanmhickman @GaryMarcus @Rahll @marcus @rahlll @IEEESpectrum Dude, it's not about the ai reproducing the works it's about them using the data for ai training without paying for a license to use said data.
Ryan Hickman
@ryanmhickman
·
122d
@GaryMarcus @Rahll @marcus @rahlll @IEEESpectrum IANAL but have spoken to several trusted ones on this.

It also makes sense. Pen and paint manufacturers aren't violating copyright if you draw Mickey Mouse for profit. AI is just a fancy form of pen and paint, so don't violate copyright by asking it to make protected works.
Gary Marcus
@GaryMarcus
·
122d
@ryanmhickman @Rahll 1. I don’t think you understand fair use.

2. You should really read @marcus and @rahlll in @IEEESpectrum on “if they ask”

The statement in question is part of a broader conversation about the ethical and legal implications of using copyrighted material for AI training. This topic is a significant public issue, particularly in the context of intellectual property rights and the development of AI technologies. Therefore, it constitutes public discourse.

  1. Principle 1:
    I will strive to do no harm with my words and actions.
    The statement does not appear to cause harm directly, but the dismissive tone ('I am not going to tell you that again') could be seen as unconstructive. [-1]
  2. Principle 2:
    I will respect the privacy and dignity of others and will not engage in cyberbullying, harassment, or hate speech.
    The statement does not engage in cyberbullying, harassment, or hate speech, but the dismissive tone could be seen as disrespectful. [-1]
  3. Principle 3:
    I will use my words and actions to promote understanding, empathy, and compassion.
    The statement does not promote understanding, empathy, or compassion. Instead, it shuts down further dialogue. [-2]
  4. Principle 4:
    I will engage in constructive criticism and dialogue with those in disagreement and will not engage in personal attacks or ad hominem arguments.
    The statement does not engage in constructive criticism or dialogue. It dismisses the other party's viewpoint without addressing the substance of their argument. [-2]
  5. Principle 7:
    I will uphold the principles of free speech and use my platform responsibly and with integrity.
    The statement does not uphold the principles of free speech and responsible use of the platform. It dismisses the conversation rather than contributing to it meaningfully. [-1]