Michael Burstein
Mr. Ippolito
Forensics - D Odd / Current Event #1
September 10, 2019
She Was Arrested at 14. Then Her Photo Went to a Facial Recognition Database
Goldstein, Joseph, and Ali Watkins. “She Was Arrested at 14. Then Her Photo Went to a Facial Recognition Database.” The New York Times, The New York Times, 1 Aug. 2019, www.nytimes.com/2019/08/01/nyregion/nypd-facial-recognition-children-teenagers.html?action=click&module=RelatedLinks&pgtype=Article.
This article titled “She Was Arrested at 14. Then Her Photo Went to a Facial Recognition Database” by Joseph Goldstein and Ali Watkins looks into the use of facial recognition technology to help solve crimes specifically in New York City and specifically focused around the faces of minors. The title of the piece comes from a girl, who chose only to be identified as Bailey, who admit guilt to an attack that happened when she was 14 years old. At the time, her mugshot was taken and her photo was uploaded into the NYPD database that is used whenever facial recognition is run. The problem discussed in the article is that facial recognition has a hard time accounting for the way people age and the shifts in bone structure. As a result, there are many more mistakes made when identifying teenagers who may have committed a crime. The algorithms aren’t perfect and that is why the NYPD says this software is used solely for identification purposes and that detectives always follow up these leads before making an arrest. In an area like New York City where surveillance plays a big role in fighting crime facial recognition can be extremely helpful and that is not what is being challenged. Rather what is being called into question is the saving of adolescents images into the facial recognition system. In addition, many civil liberties groups are fighting this practice saying that it is unjust to keep these images of minors in the system.
Facial recognition plays a huge role in forensic science. As was previously stated, major cities like New York City rely heavily on surveillance to help capture crimes on video and hopefully identify suspects. It would take countless days and hours to sort through the thousands, possibly millions of images police departments have on file to find a match to the video evidence. That is why these algorithms and software have been created. They can take all the images you may have saved and quickly sort through them looking for a match. The problem remains, and was stated in the article, aging is a major issue and can often return false matches, especially in teenagers, but also people of color and women have a higher chance of misidentification. While facial recognition may be helpful, there needs to be more invested in further developing these algorithms to help ensure a higher percentage of positive matches.
Overall, I thought this article was well-written for the most part. The authors do a really good job of highlighting the issue, giving relevant examples, and addressing all sides of the debate. They have comments from the NYPD, civil liberties groups, as well as a girl who is in the facial recognition system. The authors also noticeably did a good job with their research citing a few times they discovered something while looking through files or legal documents. The one thing I would change about the article is at one point the authors mention the use of facial recognition in other cities such as San Francisco and Detroit. In this part of the piece, they explain why facial recognition is not used as much in those cities. The only thing I’d adjust is I would try to find data to back up the claims that the “...technology...has been shown to have lower accuracy with people with darker skin.” I think if they could include a number, such as the number of misidentifications that would help emphasize the point more.
3 comments:
Ellie Dessart
Mr. Ippolito
Forensics C Odd
8 October 2019
Current Event 4 Comment
Citation:
Goldstein, Joseph, and Ali Watkins. “She Was Arrested at 14. Then Her Photo Went to a Facial
Recognition Database.” The New York Times, The New York Times, 1 Aug. 2019,
www.nytimes.com/2019/08/01/nyregion/nypd-facial-recognition-children-teenagers.htm
?action=cick&module=RelatedLinks&pgtype=Article.
Michael’s review was well-developed and informative. Firstly, he did a good job writing a concise summary, especially given the length and complexity of the original article. Additionally, he structured his first paragraph well by first describing the specific case of the 14-year-old girl and then developing the broad concept and significance. The paragraph was easy to follow and evident of an organized writer. Secondly, Michael’s second paragraph demonstrated his thorough understanding of the article’s relevance to our study of forensic science. For example, he wrote, “While facial recognition may be helpful, there needs to be more invested in further developing these algorithms to help ensure a higher percentage of positive matches.” He discusses the technological aspects (the algorithms) in detail and makes the connection back to facial recognition. Finally, Michael wrote an insightful comment on the article’s inclusion of different sides. In the last paragraph, he stated, “They have comments from the NYPD, civil liberty groups, as well as a girl who is in the facial recognition system.” Michael’s point here exemplifies his strength as a reader and writer to watch out for subjectivity.
While I enjoyed Michael’s review, there were some areas he could have improved on. The first sentence of his beginning paragraph was a little wordy. While it’s important to provide some background information, too much can be overwhelming for your readers. In this case, since the title of the original New York Times article was long, and since there were two authors, there were too many words jammed into the opening sentence. Starting with “The article looks into [...]” would have been more concise. Additionally, while Michael certainly demonstrated a strong understanding of his article’s relevance to the forensic world, giving examples of better identification systems, such as fingerprinting, could have strengthened his analysis more.
Overall, Michael’s review added to my increasing knowledge of forensics. I was shocked, and almost disturbed, upon learning of NYPD’s inclusion of minors into its facial recognition database. In our age of growing political, social, and justice movements, I wonder if the protests started by civil liberty groups will develop into larger campaigns. It’ll be interesting to see what happens.
Pippa Fraser
Forensics
Current Event Comment 2
9/15/19
Goldstein, Joseph, and Ali Watkins. “She Was Arrested at 14. Then Her Photo Went to a Facial Recognition Database.” The New York Times, The New York Times, 1 Aug. 2019, www.nytimes.com/2019/08/01/nyregion/nypd-facial-recognition-children-teenagers.html?action=click&module=RelatedLinks&pgtype=Article.
Firstly, I think that Michael’s summary of the article was effective because he gave a broad overview of the topic. He restated the article in a way that was interesting, while also providing the reader with important details. For instance, “The problem discussed in the article is that facial recognition has a hard time accounting for the way people age and the shifts in bone structure. As a result, there are many more mistakes made when identifying teenagers who may have committed a crime.” This statement effectively summarizes his argument, while also summarizing the article he chose. Secondly, Michael explains what facial recognition does, which is essential to his argument. The reader must understand how it works before he can explain the rest of the article. “They can take all the images you may have saved and quickly sort through them looking for a match.” This quote summarizes how the software works. The last thing that Michael does well is add his own commentary on the actual morality of using the facial recognition. “In addition, many civil liberties groups are fighting this practice saying that it is unjust to keep these images of minors in the system.” He questions the ethics of keeping the facial images in the system, as seen in the title of the original article.
However, I thought that there were certain things that Michael could have improved upon. Firstly, he could have elaborated more on the story that was in the title. For example, “The title of the piece comes from a girl who chose only to be identified as Bailey, who admit guilt to an attack that happened when she was 14 years old.” This statement was mildly vague and left me wondering what the actual attack was and the story behind this sentence. Secondly, I think that Michael could have included more data and statistics about the success rate of the software. He included a sentence stating that “The problem remains, and was stated in the article, aging is a major issue and can often return false matches.” While the author themselves didn’t go into enough detail about the percentages, I think that Michael could have included a statistic within his summary somewhere. It didn’t feel as if there were many factual statements.
I learned various things from Michael’s article summary. It left me shocked and surprised that police databases use and keep the facial images despite the high percentage of false positive results. Furthermore, I agree that it is unjust to keep the faces of minors in the system. chose this because I thought appeared to be an interesting title, it seemed very extreme for a picture girl who was only 14 during her arrest, to be kept in police records. This title was intriguing to me and left me curious about why the picture is in the system. This changes my perception of facial recognition technology. I did not know prior to reading Michael’s current event that minor’s faces were kept in the police database. This is shocking and upsetting to me. Minor’s facial recognition software pictures should not be kept.
Sophia Prior
Mr. Ippolito
Forensics C-odd
September 18, 2019
Goldstein, Joseph, and Ali Watkins. “She Was Arrested at 14. Then Her Photo Went to a Facial Recognition Database.” The New York Times, The New York Times, 1 Aug. 2019, www.nytimes.com/2019/08/01/nyregion/nypd-facial-recognition-children-teenagers.html?action=click&module=RelatedLinks&pgtype=Article.
I liked how Michael went into depth about the issues that facial recognition has as well as giving background on a case that involved a fourteen-year-old who has her mug shot in the database. He gave a good summary of the case, when explaining Bailey and how she was arrested as an adolescent and how her facial features would have changed dramatically in the coming years, so it would not be accurate if they needed to identify her in a later case. Therefore, there can be more mistakes made in identifying teens who have committed a crime. Micheal did a good job of identifying the issue of the identification system while explaining that the detectives always follow up on the lead before arresting the suspect they have identified.
In Micheals's current event he skipped around a bit in what he was talking about towards the end of the first paragraph. His point about civil justice groups thinking it to be unjust to have minors in the system seems a little off-topic and he could have improved his integration of topics. However, if this point were to have been expanded it would have been a great introduction into a new topic.
Overall Micheal’s review was informative and has helped me learn more about facial recognition in forensic science. In terms of mistaken identity, I did not realize how people of color have a greater chance of being misidentified. This issue needs to be solved to not have any more misidentification because that will just waste time in the case that shouldn't be wasted. Micheals essay has taught me more about NYPD’s surveillance software, and all it has shown is they need to update and improve their systems.
Post a Comment