5 ms·
This runs into the core problem with technology--we answer "What can we do" before "Should we do it" and "What are the impacts" Let's say you take your hello k
by balls187 4y ago
This runs into the core problem with technology--we answer "What can we do" before "Should we do it" and "What are the impacts"
Let's say you take your hello kitty dot art, and make a poster promoting a commercial event. You then take it to FedEx Kinkos and use a self-service copy machine to make 1000 copies. You could reasonable argue that you are violation of copyright infringement, and the photocopier / FedEx kinkos isn't.
Now instead, you have AI generate a poster, and it generates a very similar image to hello-kitty. It's arguably so similar than a reasonable person would say it's a copy. You take that poster and again make 1000 copies. Is there copyright infringement? If so, who if anyone, is liable for damages?
- airstrike 4y agoWhoever put the poster up for display and reaped some reward out of it is liable for damages. Everyone else is just doing their job in the supply chain. We want supply chains to work for the good of the economy, which is a proxy for increasing availability and reducing prices of "goods and services" to the average person.
- balls187 4y agoImagine a paying Adobe CC customer. They use Firefly to generate a poster, and unbeknownst to them, the image it generated is a reasonable facsimile of a copyrighted/trademark character. The person has inadvertently committed copyright infringement. So does Firefly need to come with a warning? The safer solution, to the chagrin of another commenter, is for Adobe to neuter the tool by only training on data in which Adobe has express permission to use.
- whatarethembits 4y agoA simple warning that what’s been generated looks similar to something that’s copyrighted is not a bad idea. Then it’s up to the AI user to do their due diligence if they intend to use the resulting work for commercial purpose. Neutering the tool from the get go is a step too far.
- codeyperson 4y agoPeople accidentally recreate other companies logos in Adobe Illustrator all the time.
- airstrike 4y agoSurely with all our contemporary AI prowess we can train a model that identifies "reasonable facsimiles of copyrighted/trademark characters" after generating them and alert the user that it could be argued as such. Still, let the user decide. We do not need creative technology to regulate observance of copyright law. (By the way I think the chagrined other commenter was yours truly ;-))
- chillfox 4y agoWith that approach you risk ending up in a very frustrating loop of copyrighted works... A bit like picking a name in an MMORPG that's been out for a few months ends up being a hell of constantly getting your name requests rejected over and over again.
- muskmusk 4y agoNot really. You just have checkbox somewhere that says "don't copy" and then it won't. Leave the decision to the user instead of baking it into the technology.
- chillfox 4y agoPretty sure "just having a checkbox" is a massively difficult problem when it comes to AI tech and baking it in is much easier.
- irrational 4y ago> This runs into the core problem with technology--we answer "What can we do" before "Should we do it" and "What are the impacts" The counter is always, “Well, we can see our way forward to doing it, which means someone else can also do it. So… we better beat them to the punch.”