In the article “These new tools let you see for yourself how biased AI image models are,” Melissa Heikkilä, argues that AI-generating image systems are applying harmful biases and stereotypes. Despite the company's attempts to fix these issues, Heikkila's research with Hugging Face and Leipzig University has created a tool to bring to light these biases. The tool highlights biases and stereotypical images based on gender and ethnicity. Heikkila blames these biases on the training data used which reflects off of American culture/values. Overall, the article clarifies these AI challenges and emphasizes making the system less biased.

I prefer this article since it gives me a detailed overview of the biases in AI image generation and highlights the efforts and challenges while attempting to confront these issues.

Tiku, N., Schaul, K., & Chen, S. Y. (2023, Nov 12). AI-generated images show a world of stereotypes. The Washington Post Retrieved from https://ezproxy.lib.uwm.edu/login?url=https://www.proquest.com/newspapers/ai-generated-images-show-world-stereotypes/docview/2888705247/se-2

Melissa H. "These new tools let you see for yourself how biased AI image models are," https://www.technologyreview.com/2023/03/22/1070167/these-news-tool-let-you-see-for-yourself-how-biased-ai-image-models-are, found by using Google.

Kalluri addresses the importance of addressing the issue of AI image biases, attempting to avoid harmful stereotypes and ensuring fair and accurate images.

Kalluri reveals that AI-generated images are often generated with biased and stereotypical images due to how they were trained. Efforts are made to reduce these biases with new data training and development to improve their algorithms. Kalluri addresses these biases since she believes AI-generated images must become widely accepted in various parts.

In "AI Image Generators Often Give Racist and Sexist Results: Can They Be Fixed?" by Ananya, the article dives deeper into sources of racial and gender bias in AI images and how to confront the issues.

Throughout this whole article we mainly focus on the newest version of AI, GPT-4, I'm just curious on the past types of AI and how it compares to the modle used today. How big is the difference and how far has AI come since GPT 1-4.

I think the big picture of the article is implying that with the prediction and language of AI start to get more complicated and complex, it gives it abilities that match humans. While people argue for this, people also argue against the idea and imply that AI is just mimicking human reasoning and pattern.

I'm using this quote again because it connects to the video on where ChatGPT models learn how to predict words for their responses. The prediction from ChatGPT leads to complex possibilities and Dobrin connects to this where context is the main reason why AI can make these predictions.

This sentence connects to Dobrin's perspective on ChatGPT limits, arguing whether ChatGPT actually understands the language or just copies answers that sound right. Even in the video, it talks about ChatGPT's model where it creates responses based on context rather that being conscious.

I think this quote is important since it connects to the subject of AI's prediction and how it could possibly positively affect the future for humans. Predicting organic ecosystems in the future for organisms can be very important and informative for the future.

This quote pretty much tells us the success of ChatGPT because of their advance prediction. Their prediction method is so advance, that their language models learn the context and adapt to knowledge given to them. Which helps with the complexity of ChatGPT.

I'm not very familiar with the word but from what I'm thinking, I think in this section it'll give us a overlook on how language models process language in a deeper or more complex level.

In this section, I think we'll learn how word vectors are transformed into different relating to sentences when their generated in ChatGPT.

Key Points: - ChatGPT utilizes the "transformer" model which was used on past AI models. What the Transformer model does is to improve the AI context of the text by focusing on different parts of a question or sentence that people put in the AI. - Token was a important part of the video, text that are submitted into the ai are broken down and turned into tokens. Tokens are then processed by ChatGPT and utilizes them for different texts, punctuation, and vocab.

Question: How important is the Tokenization in ChatGPT, would it affect the overall words that it gives us or would it just affect the vocabulary or complexity to the response?

How can educators use Gen AI effectively where the information it gives us is actually true?

Is there a way where we can identify "hallucinations" and be able to report the false data?

GANs are types of different machines, each have a generator and a discriminator. It pretty much makes new false data and makes it look like the real data.

is part of the computer that allows it to understand complex or complicated data sets. It works with pictures and even videos.

a part of AI that allows computers to adapt and learn from data/information. It improves them in doing tasks and understanding how to do something without needing a complex code or be specifically programmed to do so.

The fact that AI is able to create "Hallucinations" is startling. If AI can put up convincing information and mark it factual even when it's not could be a big problem. It shows the importance of people verifying the AI information before being used.

I found this surprising and it just proves that the problem of AI fabricating false detailed information and is a great example on why relying on AI is not a very good idea.

This quote is important since it tells us that the deep learning machine is able to copy human cognitive process, pretty much copying a human brain showing advance AI's capabilities.

I think this quote is important since it underscores the difference between traditional and machine learning. Unlike the traditional machine, machine learning allows the program to predict things without a specific programming

The section would most likely give an overview towards original type of AI and compare it to new methods of AI. Such as, the adaptive or learning type of AI.

I'll put the book in educational genre. Reading the first and second chapter is informative and educational towards AI technology and towards our education system. I expect more history about AI and how much impact they'll cause.

Question: Learning about the history of writing and how there were concerns on their upcoming, how can this be compared to the present debate towards Gen AI?

Reaction Annotation: Reading this section of the chapter changed my overall perspective on journalism, the shift of standards that most of these press have changed due to AI and it gives us a general view on what skills are required for the new generation of journalism.

Question: Based on what I read, what's a ethical method that can be taught to students for the use of AI in class or for their education?

Restatement Annotation: Dobrin used this section to reiterate his thoughts and demonstrate the gap between AI and the education system. How AI's conversion to the classroom is far beyond what school's have imagined and struggle to embrace the problems of how to instruct teacher's with AI.

Reaction Annotations: It shocks me that so many people logged in to ChatGPT, but at the same time it really doesn't since it's a new wave for students to cheat and make their lives sufficiently easier. It also surprises me that so many people knew about the program so quickly.

Restatement Annotation: Dobrin is essentially confronting the transformation of AI, Interpreting new ways to confront, navigate, and explore around these new "challenges" and how to possibly address it.

Tracking Annotation: I think the author is trying to give examples and an overview on the history of writing technologies, and how every new piece of writing brought a wave of panic to the community,.

Tracking Annotation: I expect the following passage to give a brief summary or a overall idea on what Generative AI is and how it impacts the world as of today.