AI is stealing writers’ words and jobs…

[Jfdlsjfd]

so I was looking at civitai and saw this as part of one model's description "This new model was fine-tuned using a vast collection of public domain images"

I don't think civitai would be the best place to look for, as their models are finetunes or merge of existing models, which are accused of being unethically trained. Therefore any model being built on top of that would generate the same reaction, I guess, depending on one view of what is ethical. Microsoft does claim its AI to be ethical, and who knows how Dall-E's dataset was obtained, but OpenAI is sued for its LLM model, so maybe they are not as ethical as they claim. Firefly is undoubtedy clean (because Adobe acquired the rights on the images they used), but, since training a base model from scratch is extremely expensive, there will be few effort to make a state of the art model out of only public-domain data unless there is a clear determination that it is illegal to do "business as utsual".

Edit: I forgot about pixart-alpha model being trained on CC-0 dataset: https://mpost.io/akash-networks-mainnet-8-upgrade-boosts-visibility-for-cloud-gpu-operations/

But it's more for demonstration of their traing method (which is, they say, very cost-effective, reducing the need in computation power to around 30k$)than to provide state of the art generation.

 

[trappedslider]

Good. Here’s to hoping for more.

Nightshade, a free tool to poison AI data scraping, is now available to try

Nightshade was initially introduced in October 2023 as a tool to challenge large generative AI models. Created by researchers at the University of Chicago, Nightshade is an...

www.techspot.com

so it seems some redditors have tried it and well

 

[Justice and Rule]

so it seems some redditors have tried it and well

As with a lot of AI people, they're probably a bit premature in that:

Seems like the limited dataset makes actually knowing difficult to figure out how bad it can be, especially at this juncture.

 

[trappedslider]

As with a lot of AI people, they're probably a bit premature in that:

View attachment 344597

Seems like the limited dataset makes actually knowing difficult to figure out how bad it can be, especially at this juncture.

Either way It won't be fully proven until someone trains a base model with enough, millions, of nightshade images from my understanding.

EDIT: I should point out the pervious big thing Glaze basically failed rather quickly

 

[Justice and Rule]

Either way It won't be fully proven until someone trains a base model with enough, millions, of nightshade images from my understanding.

I mean, we'll see how many it needs. It definitely seems more effective than those posts indicate, so I would not be as confident as they are that it's "dead" given that it's only just hit the scene.

 

[dragoner]

AI trying to figure out who they haven't made angry, after the Taylor Swift thing ...

 

[trappedslider]

AI trying to figure out who they haven't made angry, after the Taylor Swift thing ...

Is it the tool's fault or the person using the tool or both?

 

[Ryujin]

Hell, apparently AI is even stealing people's bodies.

 

[trappedslider]

I was already aware of the AI generation influencers but not to this degree.

 

[Scribe]

Is it the tool's fault or the person using the tool or both?

The tool is still fundamentally the problem.

Step 1. Create a narrative, a setting, where anyone can claim anything is true, and someone, somewhere will both accept it, and support you.
Step 2. Flood people with enough conflicting narratives that people are overwhelmed.
Step 3. Create tools that allow for the generation of even more content.
Step 4. Keep control of these tools and algorithms in the hands of a small group.
Step 5. Replace humans, and broadcast what you want people to think is real.

TV channels are using AI-generated presenters to read the news. The question is, will we trust them?

A start-up is developing a news service presented artificial intelligence, but will it upend the parasocial relationship between television audiences and those on screen?

www.bbc.com