- Jan 2024
-
www.eff.org www.eff.org
-
Images of women are more likely to be coded as sexual in nature than images of men in similar states of dress and activity, because of widespread cultural objectification of women in both images and its accompanying text. An AI art generator can “learn” to embody injustice and the biases of the era and culture of the training data on which it is trained.
Objectification of women as an example of AI bias
-
- Nov 2023
-
media.dltj.org media.dltj.org
-
One of the ways that, that chat G BT is very powerful is that uh if you're sufficiently educated about computers and you want to make a computer program and you can instruct uh chat G BT in what you want with enough specificity, it can write the code for you. It doesn't mean that every coder is going to be replaced by Chad GP T, but it means that a competent coder uh with an imagination can accomplish a lot more than she used to be able to, uh maybe she could do the work of five coders. Um So there's a dynamic where people who can master the technology can get a lot more done.
ChatGPT augments, not replaces
You have to know what you want to do before you can provide the prompt for the code generation.
-
- Sep 2023
-
fortune.com fortune.com
-
analyticsindiamag.com analyticsindiamag.com
-
considering that Llama-2 has open weights, it is highly likely that it will improve significantly over time.
I believe the author refers to the open-sources of llama-2 model. It allows quick and specific fine-tuning of the original big model.
-
- Jul 2023
-
arxiv.org arxiv.org
-
AI-generated content may also feed future generative models, creating a self-referentialaesthetic flywheel that could perpetuate AI-driven cultural norms. This flywheel may in turnreinforce generative AI’s aesthetics, as well as the biases these models exhibit.
AI bias becomes self-reinforcing
Does this point to a need for more diversity in AI companies? Different aesthetic/training choices leads to opportunities for more diverse output. To say nothing of identifying and segregating AI-generated output from being used i the training data of subsequent models.
Tags
Annotators
URL
-
- May 2023
-
maggieappleton.com maggieappleton.com
-
Some of these people will become even more mediocre. They will try to outsource too much cognitive work to the language model and end up replacing their critical thinking and insights with boring, predictable work. Because that’s exactly the kind of writing language models are trained to do, by definition.
If you use LLMs to improve your mediocre writing it will help. If you use it to outsource too much of your own cognitive work it will get you the bland SEO texts the LLMs were trained on and the result will be more mediocre. Greedy reductionism will get punished.
Tags
Annotators
URL
-
- Dec 2022
-
garymarcus.substack.com garymarcus.substack.com
-
every country is going to need to reconsider its policies on misinformation. It’s one thing for the occasional lie to slip through; it’s another for us all to swim in a veritable ocean of lies. In time, though it would not be a popular decision, we may have to begin to treat misinformation as we do libel, making it actionable if it is created with sufficient malice and sufficient volume.
What to do then when our government reps are already happy to perpetuate "culture wars" and empty talking points?
-
anyone skilled in the art can now replicate their recipe.
Well anyone skilled enough who has $500k for the gpu bill and access to and the means to store the corpus... So corporations I guess... Yey!
-