Freesound Forums

I finally uploaded a track just to get me started as a contributor. But I see this error next to the track:

"You haven't set your preference regarding the usage of your sounds for training generative AI models. While no preference is set, the default option will be used. To disable this warning, please set your preference in your account settings page and save profile settings."

I cannot for the life of me see where in my account / profile settings this preference is.

Can someone point me in the right direction?

These settings are at the bottom of https://freesound.org/home/edit/

FAQ entry: https://freesound.org/help/faq/#can-my-sounds-be-used-to-train-generative-artificial-intelligence-gen-ai-models

My 5ct: The text seems overwhelming. I haven't yet grasped how these additional restrictions can be added to cc, especially cc0.

PS: Not a fan of the wording "I agree". I don't understand whether this 'setting' exists to give Freesound/UPF/MTG permission or whether this information is/will be provided externally.

I feel like I'm going crazy, is it staring me in the face and I'm being old?

Yeah I just created my account and was searching for the same thing. What am I missing here? Where's the option?

Hi,

The setting should appear at the bottom of the main account settings page: https://freesound.org/home/edit/, but it is definitely not there in your screenshot. This is because you have uploaded no sounds. It only appears to users that have uploaded sounds, but the warning appears to everyone. I should make the warning only appear to users that do have uploaded sounds, I'll fix that.

The text seems overwhelming. I haven't yet grasped how these additional restrictions can be added to cc, especially cc0.

PS: Not a fan of the wording "I agree". I don't understand whether this 'setting' exists to give Freesound/UPF/MTG permission or whether this information is/will be provided externally.

If there are suggestions for simplifying the text we can take them into account, although it is not always easy because it also needs to be as specific as possible. The discussion about how this works with CC licenses can happen in this other thread: https://freesound.org/forum/freesound-project/44672/. Generally, there's no definitive answer. This is just a way through which authors can express their preference, but the legal implications are still unclear. In terms of copyright, CC does not offer a specific framework to deal with AI training, but limitations for AI training do not necessarily need to be enforced through copyright. At the moment, such preferences are an indication to model developers (not only us, this is for external researchers and companies as well). If there is a more standardized way to express such preferences in the future, we can hopefully adopt it.