What I think was the main ethical issue with the discussion on Monday was how individuals and society can be influenced by the manipulated data. There was a consensus on the most important and urgent ethical issue, which was privacy regarding personal information. This is because some people use personal information for commercial or political purposes. Moreover, there was a discussion about Deepfake videos, which are fake videos that people cannot really distinguish from the real videos with naked eyes. This artificial intelligence and deep learning technique have been already used for making pornographic against celebrities and politicians.

Examples above are both about how manipulated data deeply impact how we see understand and absorb fake things on the world. Based on our personal information, companies make different advertisement tailored to each personality, and politicians give different types of messages. This can be dangerous because it can be a new phase that polarizes our society, divided by those who have big data and those who don’t have access to big data.

Deepfake videos are also huge problems as well. We all know the famous fake video of Obama giving a speech but he did not actually. Even though deepfake vidoes are yet used politically as much as they are for making pornographic, as the deep learning and artificial intelligence get better every time, it is a matter of time that we might see fake videos that instigate, reproduce and spread out hatred and disinformation.

Nevertheless, there is not a substantial solution to these problems since they are uncontrollable and inevitable unless every person quit using the internet. Thus, each individual, especially those people who manage websites and portals, need to be more responsible.

With the Digital Age comes an increase in the amount of personal data that is stored electronically. An article by Kurt Saunders and Bruce Zucker highlights the dangers of identity theft in a time of digital repositories of personal information. Banks, schools, hospitals, and countless other businesses now store information like our social security numbers, dates of birth, and addresses in online databases that can be hacked. Grinnell College students experienced a social security number hack first hand this year! Digitized information makes it easier for thieves to assume someone else’s identity and withdraw money from their bank account or take out credit cards in their name without the intention of ever paying the money back. As such, the question is, just how can such a matter affect a person, and how is it being combatted.

There are two practical issues related to identity theft. A person’s identity is a fundamental component of financial institutions; banks don’t give loans to those unlikely to repay them. A person can open lines of credit with no intent to repay under a false identity, thus gaining access to otherwise unattainable resources and damaging the victim’s ability to do the same. The real-world case of Terry Rogan, a man from Michigan, depicts issues related to law enforcement; Rogan’s identity was assumed by a fleeing convict, who later had an arrest warrant for murder issued in Los Angeles. The real Rogan’s information was placed in a national database, and every subsequent interaction he had with law-enforcement back in Michigan was governed by that. Rogan was detained several times on suspicion of a murder he didn’t commit after otherwise routine traffic stops.  This scenario is not too far fetched for anyone, and brings the question, on just what is being done to counteract potential threats.

Due to identity theft becoming a real threat to almost anyone in the Digital Age, repercussions have been made in order to keep people and their information safe. In what was necessary, the government initially did not have laws that would qualify online identity theft as a criminal offense. Originally, they could only utilize laws that made unpermitted payments and fraud illegal, but this failed to reach the extent of being able to proactively combat against the growing threat of identity theft. Eventually in 1998, the Identity Theft and Assumption Deterrence Act was passed. This act allowed for identity theft to be combatted against. First, it allowed for the government to charge against online identity theft as a criminal offense with severe prosecution against perpetrators, and the act also allowed more people to become more aware of online safety due to federal regulations to educate people in schools of the dangers of the web. While it’s hard to say if such actions will truly prevent people from suffering identity theft, it’s a start towards combatting against the crime, as people recognize it’s severity to the victims and the public, as well as prompting them to take a stand against such crimes in order to combat against it.

Deepfake videos utilize artificial intelligence and machine learning algorithms to superimpose the faces of famous people, or even ordinary folks, onto the bodies of other people in different videos. Deepfake videos appear very realistic because machine learning allows the algorithm to continuously improve the image. This technology is often used in a sinister way to incriminate people in pornographic videos or to generate visual ‘evidence’ of events that never actually happened. The open-source artificial intelligence tool, TenserFlow, has been misused as a tool for creating Deepfake videos as the program relies on machine learning and image processing.

“Deepfake technology does its own google image search and scours through social media and can, by itself, replace faces in videos.  The program improves itself independently through machine learning.  Using this tech, anyone can create fake videos including pornographic videos of just about anyone.  Also, Deepfakes can be spread rapidly considering how quickly media is consumed and reproduced online. There is another technology that lies in this same vein, such as tech that can automatically alter images, and tech that can recreate voices.  The implications of this are that we may be looking at a future where people can create photos, videos, and audio of someone doing things they never did and saying things they never said.”

Development of crafty fake videos techniques influences jurisdictive decisions in courts as well. Videos and images have been strong evidence to prove eyewitnesses, but now it came to the situation that the validity of videos and images may be skeptical. Rather, videos might need to be proved by eyewitnesses. Another problem is that, yet, there are not many experts who know well about artificial intelligence and are capable of distinguishing fake videos from the real videos.

While it is only a matter of time before fake videos keep improving their performance, research in this field is sparse and so too are the necessary tools that could help stop the perpetuation of deep fake videos. Researchers have begun to make algorithms that can make these deep fake videos more easily identified but this same technology can also further advance the creation of these videos.

Sean Lee: Third paragraph, editing, and posting

Zaria: First paragraph

Gray: Second paragraph

TJ: Fourth paragraph

Northpointe’s algorithm takes real people, with complex identities and lives, and reduces them to abstract equations. The job of a judge is to listen to the evidence presented by the defence and the prosecution, consider the whole picture, and issue a decision based on the available facts and the the law. When something like a mandatory minimum is introduced into the legal system, it takes part of the process of thinking and adjudication away from the judicial process. It is certainly true that the biases and prejudices of our greater american public are already reflected in our judges. When Northpointe’s algorithm, however, is used in matters of jurisprudence to influence bond and sentencing two things are occurring. First, we are relying on computers and programs to do our thinking for us and pretending that a number can accurately represent the complexities of a human life. Second, we are obscuring the biased process which gives a false notion of objectivity.

I wonder how ethicists, sociologists, lawmakers, and community members were consulted to determine what questions should be asked and how they should be weighted? Even seemingly innocuous questions such as where someone lives are tied up in issues of racial and economic segregation in America.