foregoneconclusion

JamesCude said:

Oooof hope they change that title. Cold as ice.

'Mommie Dearest' was already taken.

Globally, it looks like legislators think App Store commissions are the big bad wolf and crypto/AI are warm, fuzzy bunnies. Bizarre at best. 

randominternetperson said:

foregoneconclusion said:

mfryd said: I am not a lawyer, however this is my understanding of what someone can do without violating copyright law.

Yes, a person can do that without violating copyright. But AI doesn't work like the human mind. AI requires the complete works of Stephen King to be copied into a database. If it's done without permission, then it's a violation of copyright. 

Not according to this week's ruling.

"In short, the purpose and character of using copyrighted works to train LLMs to generate new text was quintessentially transformative.  Like any reader aspiring to be a writer, Anthropic’s LLMs trained upon works not to race ahead and replicate or supplant them — but to turn a hard corner and create something different.  If this training process reasonably required making copies within the LLM or otherwise, those copies were engaged in a transformative use."  

https://storage.courtlistener.com/recap/gov.uscourts.cand.434709/gov.uscourts.cand.434709.231.0_2.pdf page 13

Yes, but the judge is ignoring the fact that the AI program is a professional product and a human being is not. Plus, if the AI is really all that "transformative", why wouldn't AI companies simply restrict the training model to public domain material? Answer: they know that it isn't really as "transformative" as the judge thinks it is. For example, if the AI was only trained on the text from Bazooka Joe comics then take a wild guess at what the output would be like? Not all that "transformative". 

mfryd said: I am not a lawyer, however this is my understanding of what someone can do without violating copyright law.

Yes, a person can do that without violating copyright. But AI doesn't work like the human mind. AI requires the complete works of Stephen King to be copied into a database. If it's done without permission, then it's a violation of copyright. 

mfryd said: I am merely pointing out that these are a complicated issues.

"Fair use" isn't really complicated. I'll provide an example that applies to myself. I like to design my own graphics for movies and TV shows that I have ripped from discs and stream from a home media server. To create the graphics, I typically go to the internet and look at a variety of posters and packaging that exist for that particular movie/show and choose the images and typographic treatments I like best and then digitally edit those elements into a new composition. This approach is "fair use" ONLY because the copyrighted material that I'm using without permission will not appear anywhere other than on my home Apple TV. In other words, it's purely for personal use. It has no commercial or professional application. And I'm not displaying these graphics to the general public.

However, if I put those same graphic treatments into a professional portfolio to try and get a job designing those kinds of graphics, it wouldn't be "fair use" anymore. I would be violating copyright because I had never received permission to use any of the material professionally. 

So you can see how the ruling by this particular judge is ignoring a very obvious copyright issue in regards to permissions. The only way an AI program that was trained on copyrighted material without the appropriate permissions could be considered "fair use" would be if the AI program was never made available to the public. Literally like if Sam Altman was the only person that could use ChatGPT. Because once it's publicly available as a product of a professional organization, it can't possibly be considered personal use anymore...just like the example of putting my home ATV graphics that used copyrighted material without permission into a professional portfolio.