This info is current with the latest AI releasted and with the app's position as number one, it's relevant because it is an open source AI rising in charts over the current closed source AI, Authority is decent because the verge is reputable and information is verifiable, except for the "By contrast, OpenAI CEO Sam Altman has said GPT-4 cost over $100 million to train." quote, which is behind a paywall, but if I verify this, combined with another recent piece of evidence, stargate, will be enough to show just how much money is beind dumped into AI, and combined with some benchmarks, I can prove how little the difference open source and closed source AI's are in quality.

Softbank, O. (n.d.). Announcing the stargate project. openAI. https://openai.com/index/announcing-the-stargate-project

This info is current, it was published january 21 and nothing new has come up, it is relevant because it is pouring billions more dollars into closed source AI (while still being unprofitable), it's authority is doubtable because the company announcing it is the one doing it, which means that they may twist the truth to make it look better, but the information is verifiable. I want to use "The Stargate Project is a new company which intends to invest $500 billion over the next four years building new AI infrastructure for OpenAI in the United States. We will begin deploying $100 billion immediately" to illustrate just how ridiculous and massive is the amount of money being poured by the bucketloads into AI.

The article talks about how nations across the world are racing to find a way to responsibly regulate AI and Machine Learning software. They look at how they can make sure that innovation can still prosper while still doing things responsibly. The article is credible because it shows the many ways that can use AI to find the right balance to breathe in the race for AI. Besides minimizing risks, such regulation could facilitate AI’s uptake, boost legal certainty, and hence also contribute to advancing countries’ position in the race. 

This info is the most current one because this is where they are hosting and mainting their open source product, it's relevant because it is the newest and most impressive open source AI model as of now, it's authority is again doubtable because they could stand to benefit from inflating their numbers, but they are also funded by a hedge fund dedicated specifically to advancing technology, so it's unlikely that they would do that, their "evidence" are benchmarks that I would love to use to demonstrate really how close open source and closed source models are, that with the previous financial evidence may make a case that open source AI catches up too quickly to closed source for them to make a profit.

This source is relevant because it's about company policy on how they handle AI, it's current with the newest policy changes at meta, accurate and trustworthy because Meta themselves wrote it, which is the company making their own policy changes.

"as well as the information that people have shared publicly on Meta’s products and services. This information includes things like public posts or public photos and their captions." This is the most important part and what I plan to use, this is where meta says that they will be using public posts that anyone makes to train their AI models.

I plan to use this source to illustrate how companies are taking services you already use and adding privacy invading policies to train their AI models, I want to use this to eventually lead to the point of how they are making AI inescapable and therefore invading your right to privacy because they are exploiting your familiarity and perhaps dependence on existing apps.

This article is almost as current as it can be, it just happened and was published jan 29th, it is  also very relevant because it talks about deepseek, the open source AI that disturbed the industry very recently, it’s accuracy and authority is a little doubtable because it’s evidence is “OpenAI told the Financial Times”, and the link is blocked behind a paywall. I will use this source to explain distillation and how it’s a threat to spending money on closed source models.

“L.L.M.s string together words based on probability, not a deep conscious appreciation of what these words mean. For this reason, some computer scientists call L.L.M.s “stochastic parrots.” (Bloom 2) The article describes that AI and Large Language Models don’t understand concepts like morality. This is often described as LLMs being "stochastic parrots”. These parrots are described as stringing words together without any real understanding. This limits their ability to make moral decisions, as they lack the reasoning that comes with humanity. Since AI doesn’t "understand" the content it generates, it can’t engage in real moral reasoning. It replicates patterns—negative or positive—which can lead to errors or ethical issues. The currency of this article is relevant as it was published in November 2023. It’s relevant to my argument as it describes the underlying issue with AI, mentioning how AI doesn't have a conscience. The author is Paul Bloom, a Canadian-American psychologist for the New Yorker. The information is accurate as the author analyzes work from AI researcher Atari. This article aims to educate and inform, as it provides information on why AI is misinformative. 

“In three studies (combined N = 200), we tested the hypothesis that people will find it less plausible that AI is capable of judging the morality conveyed by a behavior compared to its competence.” (Oliviera 2) The quote from Oliveira's study shows that people are likelier to determine AI's ability to assess competence rather than morality in human behaviors. This observation is relevant to discussions about ethical boundaries in AI, as it shows debate over technological advancement as opposed to humanity. Additionally, it provides quantitative data which helps give a clear contrast between AI’s judgment morality. The currency of the article is relevant as it was published in 2024. The quantitative data helps augment my argument by providing clear statistics about my main idea. The authors are surveyors from the University of Eindhoven who conducted said survey. The information presented is accurate since it references the survey conducted, and the purpose of the information is to educate since it was published in an academic journal to be presented to the university.

“The need to apply ethics to AI for engineers and other technical professionals that might not have received any training studied ethics or fairness, or been given specific guidelines, is an important point.” (Rao 1) This quote strengthens my argument that AI development is not only a technical challenge but also a moral and social one. Without ethical training, the resulting systems may not only reflect personal or organizational biases but could also perpetuate historical injustices, reinforce stereotypes, or create new inequalities. Since AI systems are bringing in systems that can create severe effects in the outside world, engineers should implement algorithms through the lenses of social values such as fairness, privacy, and justice. The article was released in 2020, so the currency is relevant. This source is relevant to my argument since it provides specific examples of what AI is capable of. The author of this article is Danielle Hoffman, the director of philanthropic engagement at Carnegie Mellon University. Furthermore, the author uses Ethos to invoke credibility by citing Vincent Conitzer, an AI specialist and ethicist at Carnegie Mellon. The purpose of this article is to inform, as it was published in the college’s magazine.

Knight, Will. “The Dark Secret at the Heart of AI.” MIT Technology Review, 11 Apr. 2017, www.technologyreview.com/2017/04/11/5113/the-dark-secret-at-the-heart-of-ai/ 

"The workings of any machine-learning technology are inherently more opaque, even to computer scientists, than a hand-coded system. This is not to say that all future AI techniques will be equally unknowable. But by its nature, deep learning is a particularly dark black box." (Knight 1) This quote shows a crucial flaw with AI systems–their decision-making processes are hidden. This lack of transparency raises ethical questions in social media, where algorithms already control information catering. Without transparency, it is difficult to understand how and if AI can engage in public discourse. This article was published in 2017, meaning the currency of the topic is affirmative. Knight’s article helps augment my writing as it exposes one of the key reasons AI is untrustable–its opacity. Will Knight is a senior editor at the MIT Press and cites studies in the field to help support their argument. The article was written to inform the general public on the uses of AI.

“Despite these benefits, processing SM data is not trivial. The various types of content available on SM platforms such as text messages, images, and videos, are considered noisy and less formal compared to standard Web data, such as news articles or images published with those articles. Therefore, data analysis techniques trained and built for standard Web data often do not work well on SM data.” ( Muhammad 2) Muhammad highlights a key point regarding AI–the informal and unstructured format of social media poses challenges for pattern-making strategies made by AI. When AI misinterprets slang or context, it may start spreading incorrect information. Since the study was published in 2020, its currency is relevant. This source helps support my argument as it explicitly states how the structure of Social Media directly challenges the fundamentals of AI technology. The study was conducted by a group of researchers from around the world, with credible data provided in the form of statistics. Additionally, they cite other works in the field to assure accuracy. The purpose of this study is to gather data, as it was a public experiment.

Azgin, Bilge, and Sevki Kiralp. “Surveillance, Disinformation, and Legislative Measures in the 21st Century: AI, Social Media, and the Future of Democracies.” Social Sciences (2076-0760), vol. 13, no. 10, Oct. 2024, p. 510. EBSCOhost, https://doi.org/10.3390/socsci13100510.

“AI is frequently utilized in both democracies and authoritarian regimes for purposes of data collection on citizens and their manipulation through disinformation” (Azgin 2) This quote helps highlight the ethical concerns behind AI’s role in spreading misinformation to help push political agendas. The article was published in 2024, meaning the currency is appropriate. This source is relevant to my argument as it explicitly states how AI can be weaponized, a topic I touched upon but needed more sources to help support my idea. The authors are professors from Cyprus International University and analyze data and/or trends in data to conclude. The purpose of this article is to educate, as it was written in an academic journal to be published in Social Sciences.