(Lynch 2024)

Background

In a modern society that uses flexible and widely big data, algorithms have thoroughly infiltrated daily life, and the applications are ubiquity existing, not limited to using social media to contact others and using leisure and entertainment platforms.

Algorithms could drive the development of this digital society, for instance, in the medical area, people could use algorithms to diagnose the general symptoms and how to optimize the treatment of the patient’s program, going through the algorithmic analysis of all aspects of the data corresponds to the most appropriate medical strategy, which improves the efficiency of the health care industry, the patient will be allowed to choose their own needs independently; additionally, there is valuable to point out the advantages that algorithms bring to the area of transportation could be through the GPS satellite positioning in accordance with the traffic flow to determine whether the road is congested, and then recommended to the navigation user different needs of the navigation route planning, there is more time-saving and to avoid the occurrence of accidents; focusing on social media, the filter bubble feature brings a personalized mode of selective prioritization of the content recommended to eliminate the situation of overloaded content, while bringing positive impacts to both the user and the platform, that the platform can enhance Internet engagement through the user’s enjoyable experience and increase the time spent on the platform. Moreover, content filtering and sentiment analysis are adequately utilized to link users with social media platforms and evoke empathy (Holone 2016).

Social impact of algorithmic governance

In relation to the social aspects of algorithmic governance, Issar, and Aneesh (2022) mentioned that the vital area of concern is the problem of power, which refers to surveillance, the monitoring of users’ usage trajectories as well as leading to loss of privacy right, rising to information leakage and disenfranchisement of users, while at the same time, the problem of bias also exists. Algorithmic bias leads to unethical problems, discriminatory algorithms can lead to the limitation of users’ thoughts, long-term control by algorithmic governance is unfair to users so that social inequality cannot be slowly eliminated, and it might cause the marginalization of specific groups of people. However, algorithmic governance also has a specific dehumanizing and violent impact, and algorithms will have a particular adverse effect on human life and thus submerge some sense of unfairness into the subjective consciousness of human beings, intensifying the neglect of all aspects of human life for the problem of discrimination, which makes the asymmetry of society even more severe.

Flew (2021) also regarded a loophole in the algorithmic regulatory frameworks behind the digital social media platforms; even if the algorithmic possessed a larger amount of data that the system would have more accurate predictions, the issue of privacy and data security could not be ignored. The influence and concerns of algorithms in social media are worth exploring, and this blog post will analyze the positive and negative impacts of algorithmic governance in present society through the definition of algorithms as well as by combining them with examples that are easy for the public to understand.

Algorithms power and related issues

Influence on decision-making

The primary purpose of the principle of algorithms is to solve a specific problem and execute the corresponding instructions. Basically, algorithms have a high degree of accuracy in predicting the behavior of netizens and implicitly making decisions for them. The algorithmic governance model is “designing” the user’s mind; there are both positive and negative social impacts existing, and getting through the analysis of the algorithmic system will be done by the users’ behavioral tracking services to present customized content. In contrast, the machine projected is not the actual user’s preferences, but it will also be mixed with some commercial purposes. An algorithmic decision-making system cannot guarantee the users’ specific behaviors after receiving the information results from the algorithmic system; thus, there is an inability to guarantee the correctness of the user’s decision-making. At the same time, this is the challenge that humans should face when the algorithm is in-depth in modern society. There is also a certain risk of being misled in the acceptance of the algorithm to bring convenience. All the outcomes of the implementation of the conclusion are based on the analysis of other people or other cases and cannot be considerable decision-making results on the individual case, so that there are certain limitations with the user’s individual right to know information and decision-making, that algorithmic system might make flawed decisions, and harming marginalized populations (Ananya 2023).

(Komarraju 2021)

Algorithmic bias

In fact, algorithmic bias has left traces in all fields, and human bias towards specific groups is formed unconsciously, which may lead to the amplification of these stereotypes in the algorithm. Due to the algorithm being an analysis of fixed data, it cannot be randomly adjusted with the diversity of human beings, so using representative historical data as an algorithm will also lead to a bias in the final decision outcome towards a specific result, resulting in a bias effect.

“Algorithms are not objective, but rather embody the value-laden view that some performance is better or more important than others.”

——Sina Fazelpour & David Danks

For instance, algorithmic bias will reflect some social problems, such as gender stereotypes and even gender discrimination. Lee, Resnick, and Barton (2019) mentioned that in the data collection, the different industries will involuntarily correspond to the corresponding suitable for engaging in the gender group in the vocabulary associations or fixed occupational associations, while Princeton University researchers using AI machine learning found that when they were searching for different names, the field of art will be more associated with female, but if the programmer, science and math and other related industries or information will be considered more inseparable from male, learning from such algorithms will reinforce the subconscious gender bias of human beings.

(Cooper 2021)

Transparency and privacy issue

Transparent algorithms can establish confidence between users and social media platforms and provide accountability by allowing the identification of risk and unethical behavior (Issar and Aneesh, 2022). The algorithm operation behind the visualization can cultivate users’ trust in the algorithm system, and it could urge its execution to avoid some discriminatory results better. Such transparency permits users to understand how their private information is being used, enabling them to make informed decisions about whether to share data, and transparent algorithms help organizations demonstrate compliance with legal requirements, ethical standards, and industry best practices, thereby reducing the risk of regulatory violations and ethical dilemmas. Afterward, building integrity and trustworthiness can promote the improvement of the user’s right to know the underlying algorithm system.

How AI and algorithm actuate TikTok social media platform

User’ experiences and satisfaction

TikTok is a social media platform focused on creating and watching short videos. To bring a better user experience to content, the user needs to be entertained and obtain information. TikTok sets a personalized recommendation algorithm. However, this is achieved by using AI algorithms to analyze the user’s behavioral data and interest preferences. These data included browsing history, liking preference, interactive comments, etc. Applying these data allows TikTok to accurately recommend preferred content to users (D’Souza 2024).

(Smith 2021)

The features of content recognition and hashtag category

TikTok launched a function that can recognize content and categorize tags. It can be realized that TikTok has a very high degree of accuracy in using AI technology to identify and classify video content that will connect to the corresponding hot hashtag. On the other hand, the AI algorithm can automatically recommend that the user may use the background music in accordance with the content of the video when users are in the process of filming or posting videos. TikTok’s AI technology can also recognize the scene in the video, such as whether it is focused on the characters or the scenery; if the video appears to be a cityscape or seaside scenery, TikTok will recognize these types of scenes and recommend relevant themes or geographic locations for the user automatically. Another exciting aspect is that TikTok’s special effects creators can post various effects to entertain users while filming. AI technology recognizes specific movements and expressions to trigger a wide range of quirky effects. These AI technologies allow the platform to understand better and analyze user habits to create a personalized user experience, and the precise content interaction function provides TikTok content creators with the opportunity to showcase and promote their content.

(J.Zotara 2022)

Commercial realization conduct

Meanwhile, TikTok’s commercial realization and algorithms are closely related; the analysis of user behavioral data can predict the user’s consumption intention, depending on the likes or browsing conduct will deduce that the user’s interest in a particular product or service, and then give the user targeted advertising that improves propaganda effect and conversion rate. On the other hand, the platform also utilizes algorithms to give content creators a development space; AI algorithms identify high-quality video content automatically since the length of time and interaction rate on the video to determine or audit the popularity of its content, then indicate this video to provide more exposure and business opportunities. Simultaneously, TikTok provides ‘creator benefits’ to stimulate more superior creative content.

Struggles and concerns

Despite the convenience that TikTok’s algorithmic technology brings to users, there are still some controversies and concerns. Owing to the premise that TikTok’s personalized recommendations require users’ data, there is a risk of abusing the security of users’ data. TikTok’s AI technology involves complex algorithms and data processing processes, so a lack of transparency and regulation may lead to abuse and misconduct. In order to ensure that the use of AI technology on the TikTok complies with laws, regulations and ethical standards, users is necessary to understand the operations behind, and then hand over their personal information to protect their rights and interests. Lynch (2024) discussed that TikTok’s addictive algorithmic setup does not take adequate measures to safeguard the mental health of young people. Content is too broad to target content filtering.

Conclusion

To sum up, algorithms are now an essential component of contemporary life and have a significant impact on every facet of our everyday existence. It is difficult to dispute the effect of algorithms on social media, banking, healthcare, and transportation. This blog examines the benefits and drawbacks of many areas and how algorithms have influenced the digital world. Algorithms are becoming more common in society. However, this comes with specific difficulties. It is important to take privacy problems, data security, and algorithmic bias seriously and to solve them.

Algorithms on social media sites control what users can see, dominate their choices, and subtly affect people’s ideological stereotypes. In addition to making life easier for humans, we should make every effort to allay worries about justice and fairness brought on by algorithmic biases, particularly those that may result from prejudice and bias against particular persons or groups. In an algorithm-driven world, people have to fully utilize the conveniences algorithms offer while taking precautions against any potential risks, using objective human thinking models to improve the algorithm system for correction appropriately, and bolstering the network environment and data collection supervision to utilize algorithm governance better.

References

Holone H. (2016). The filter bubble and its effect on online personal health information. Croatian medical journal, 57(3), 298–301. https://doi.org/10.3325/cmj.2016.57.298

Issar, S. & Aneesh, A. (2022). What is algorithmic governance?, Sociology Compass. https://compass.onlinelibrary.wiley.com/doi/epdf/10.1111/soc4.12955

Flew, T. (2021). Regulating Platforms. Cambridge: Polity, pp. 79-86

Fazelpour, S., & Danks, D. (2021). Algorithmic bias: Senses, sources, solutions. Compass. https://doi.org/10.1111/phc3.12760

Barton, G., Lee, N. T., & Resnick, P. (2019). Algorithmic bias detection and mitigation: Best practices and policies to reduce consumer harms. Retrieved April 07, 2024, Brookings, from https://www.brookings.edu/articles/algorithmic-bias-detection-and-mitigation-best-practices-and-policies-to-reduce-consumer-harms/

D’Souza, D. (2024). TikTok: What It Is, How It Works, and Why It’s Popular. Retrieved April 07, 2024, Investopedia, from https://www.investopedia.com/what-is-tiktok-4588933

Annanya. (2024). Algorithms Are Making Important Decisions. What Could Possibly Go Wrong?. Scientific American. https://www.scientificamerican.com/article/algorithms-are-making-important-decisions-what-could-possibly-go-wrong/

Lynch, M. (2024). The Impact of AI on TikTok: How Algorithmic AI is and Will Continue to Impact User Experience and Content Creation. LinkedIn. https://www.linkedin.com/pulse/impact-ai-tiktok-how-algorithmic-continue-user-experience-lynch-eh5he/

Images

Lynch, M. (2024). The Impact of AI on TikTok: How Algorithmic AI is and Will Continue to Impact User Experience and Content Creation[Image]. LinkedIn. https://www.linkedin.com/pulse/impact-ai-tiktok-how-algorithmic-continue-user-experience-lynch-eh5he/

Komarraju, A. (2021). Did Humans Fail Against An Algorithm In Decision Making? [Image]. Analytics Insight. https://www.analyticsinsight.net/did-humans-fail-against-an-algorithm-in-decision-making/

Cooper, K. (2023). The AI Community Is Demanding More Transparent Algorithms—Here's Why. [Image]. Springboard. https://www.springboard.com/blog/data-science/algorithmic-transparency/

Zotara, J. J. (2022). 10 TikTok Marketing Tips & Best Practices [Image]. Search Engine Journal. https://www.searchenginejournal.com/tiktok-marketing-best-practices/437174/

Smith, B. (2021). How TikTok Reads Your Mind [Image]. The New York Times. https://www.nytimes.com/2021/12/05/business/media/tiktok-algorithm.html