Artificial intelligence has become more and more prominent in our lives. While some may not actively use AI technology, there is a high chance that they have encountered it at some point. AI has become essential in many industries, such as health care, banking institutions, retail, and business corporations. According to an article from The Harvard Gazette, private companies have been using AI to make serious decisions regarding health and medicine, employment, creditworthiness, and criminal justice. Why is it an issue if so many major companies and industries are using AI technology? The simple answer to the question is ethics. Artificial intelligence has many benefits and is quite convenient, so it has become prevalent everywhere. However, with the increase in AI popularity, the ethical issues of AI technology have become increasingly apparent, which include bias, regulation, and plagiarism.
Bias
As stated by UNESCO, “AI-systems deliver biased results.” This is because the large sets that AI uses to deliver these results are already biased. For instance, ChatGPT gets its data from the web, also known as web scraping, which uses search engine technology. The issue is that this technology is not neutral, and it relies on an algorithm based on the number of clicks the website has. Because of this algorithm that infiltrates the AI system, it can lead the software to uphold widespread biases and further push forward certain prejudices and stereotypes. Some of the most prevalent biases we see in search engine technology, and consequently in AI technology, are gender bias and racial bias.
However, biases have always existed in our world. People in the banking industry, business industry, retail industry, and even the medical industry have made significant decisions according to their biases. Some may argue that AI minimizes these occurrences and prevents favoritism. An example of such use is resume-screening software, which requires less human involvement in screening applicants. Yet, AI also has such biases. According to Michael Sandel, a political philosopher, the concern is that decisions made by AI are claimed to be objective. AI results make it seem that its judgments and predictions are objective when it is clearly not because AI replicates human biases. Mainly because humans do the programming and data input to the AI software. One must not forget that artificial intelligence is a product of humans.
Oversight (Transparency, Data Privacy)
Some would think that with how much access AI has in our daily lives, there would have been more laws regulating the use of AI. However, private companies can use AI software to make decisions affecting virtually everyone in the United States without U.S. government oversight. So far, companies that integrate AI into their system are self-policing. They don’t use any current laws regarding AI software because there isn’t any federal legislation for AI regulation. Instead, companies and institutions that use AI rely on existing laws that touch upon aspects of AI. This lack of regulation applies to the United States government. According to Victor Li, the assistant managing editor of the American Bar Association Journal, other countries have already taken steps to regulate AI software. The European Union plans to pass the AI Act, and China has drafted several rules to regulate AI.
The importance of regulation in AI systems is related to transparency and data privacy. According to Stahl’s research journal, Ethical Issues of AI, artificial intelligence “poses several risks to data protection.” Data is something all AI systems use for training purposes. For an AI to function adequately, it will need a large amount of data to learn, but access to that data can be questionable. Are the data sourced ethically? Is it safe for AI to have such information? How protected is the data, and is it safe from cybersecurity attacks? AI can be intelligent enough to detect patterns and formulate insights without too much access to personal data. This is not only an invasion of privacy but can be a severe security risk. Moreover, there is a lack of transparency with AI programs. The questions regarding data privacy are difficult to answer because AI companies aren’t transparent with their information. The public doesn’t have access to the systems, and since there is no governmental regulation, the US government doesn’t have information about these companies either. Whether AI companies are programming these AI software ethically is uncertain. With the lack of transparency, there is also a lack of accountability. It makes it difficult for people to question and address the issues of bias and discrimination with these companies if they don’t offer information on how the data is being used for machine learning.
Art (Plagiarism)
Artificial Intelligence art is a major issue in itself. It has become something everyone who is on the Internet talks about. But is it just only art? No, because the main problem is more profound than AI art; it is giving credit where it is due. Who is to be credited for art made by AI software? The original works that AI used to make its piece? The programmer behind the AI software? Or can we credit the AI itself? The questions are still not answered. It could be that this can be answered after more analysis, or it ends up being a philosophical question. Some may be confused about how this is a philosophical question, but it is because this issue involves creativity. Art, stories, and music are creative expressions for humans. Thus, it is philosophical because we are asking whether AI technology can be credited for creativity, which society assumes is a living trait. It is a debated question where an answer may not be found.
However, AI art is an ethical issue not just for who gets credited but also the aspect of plagiarism. AI takes data from other sources to come up with its results. When making AI art, the software takes small parts of various pieces to make one artwork. Essentially, AI software steals other artwork to make its own art without crediting the original artists. Currently, numerous AI art generators and self-proclaimed AI artists use original paintings and ignore that they are plagiarizing. Many artists who have put their artwork online are frustrated that it is being used without permission. In fact, some of these AI artworks get more attention than the original works, which shows how unethical it is (The New York Times). AI technology is taking away the chances for human artists to gain recognition for their talents and hard work. This particular issue is one of the more popular ongoing debates regarding the harmful effects of AI technology.
Thoughts from the Masses
A recent survey was done by the Pew Research Center about the opinions of adults in the United States regarding artificial intelligence. One question from that survey is how they feel about the increased use of AI in their daily life. As shown clearly in Figure 1, in 2023, 52% of Americans felt more concerned about the increased use of AI. And 10% of the participants felt excited. It is interesting to note how the percentage of concern increased in the past three years, and the percentage of excitement decreased. This may be due to how AI has become more prevalent today compared to in 2021. The speed at how widespread AI technology has become is a cause of concern.
Additionally, the applications of AI have mixed opinions from the respondents. Americans are very concerned about data privacy. As discussed in the Oversight section, data privacy is one of the major ethical issues of AI technology. Because there is a lack of governmental oversight, there is less protection for personal information. And if people have more public lives, it is much easier for AI technology to be used to get that information.
Conclusion
Artificial Intelligence will continue to grow and become more advanced. Some robots talk and deliver food. AI is integrated into many apps, such as Spotify, Grammarly, and Photoshop. In the near future, AI technology will become something everyone has experienced at least once. That is why being informed about AI technology is crucial. AI technology has its evident benefits, but there are also several issues regarding AI. Compared to how long technology has existed, AI has been a very recent event, so there will be flaws, whether that be discriminatory issues, lack of regulation, or copyright issues. In the beginning, AI was a novel idea, but after more experience with such technology, ethical issues are bound to arise. People are sure to question, and it is about time that such issues are addressed and resolved.
Ever since AI started getting popular earlier this year, I've always been weary of it. I'm kind of glad that the general consensus is that we're still pretty cautious of AI. It's really uncool that so many artworks have been taken unwillingly. I feel like the most important thing about art is that its a reflection of the creator, so I don't consider AI art to be real art because there is no creator. Awesome blog post, it was super informative!
I both love and fear AI because on the fun side you can play with around with filters, but for the most part I've seen people use it in a bad way as you said. Not too long ago I actually saw how the voice actress for Daphne from Scooby Doo, Grey DeLisle, was scammed about $500. It seemed she paid an artist, and when she posted the artwork Twitter or X now, a lot of people commented that it was AI which disappointed her.
Artificial Intelligence will most certainly grow within the future, rather that would be good or bad has yet to be seen. AI could get us in big trouble like you had said. There could be a chance that AI truly takes over in the future too. I never realized how big of a problem AI could really be until now.
You raise several important concerns here. I'm particularly drawn to the ways that AI is shaping/affecting art and other creative industries. There are so many issues tangled up—authorship, ownership, copyright, distribution, and more. It's a lot to consider, and I appreciate your starting the conversation!
AI has always freaked me out and I think that we should be concerned about AI. Has anyone else watched Terminator?!