The innovative teaching method was chosen by a high school computer science teacher at the Model High School for Girls in the Bronx, New York, by asking his students to study the weaknesses of the systems. artificial intelligence. In her experiment, Professor Marisa Shuman included an artificial intelligence application, a chatbot. ChatGPT, designed to provide answers to user questions, develop a course without human intervention. The ChatGPT algorithm will solve the topic of the day and create a lesson plan on technology portability.

“I don’t care if you don’t learn anything about wearable technology today. I want to view ChatGPT. The goal is to decide whether the course was effective or not,” Ms Schuman said. In the US, schools and universities are trying to manage the phenomenon of “chatbots,” artificial intelligence applications that can autonomously generate text and images that look like they were created by a human hand.

While many schools are quick to ban chatbots, believing they can be used as a “vacuum cleaner” in exams, teachers like Ms. Schumann are choosing to use technology advantages to identify potential side effects of new media. Ms Schuman says she is committed to educating the next generation of creators and consumers in IT critical thinking. This approach considers the critical study of algorithms as equally, if not more, important than knowledge of computer programming.

The New York High School District, the largest in the US with 900,000 students, trains computer science teachers to help students recognize AI biases and the risks they pose. One of the lessons includes a discussion of the failures of face recognition algorithms that recognize white faces much more accurately than black faces.

Many schools teach that powerful artificial intelligence systems are capable of creating fake news, such as realistic videos of famous political figures making statements they never made. Understanding the capabilities of such algorithms is necessary for active citizens of the future, teachers say. “It is very important that students understand how these systems work as these algorithms learn from the digital behavior of young people. AI makes a lot of decisions on behalf of young people, even if they don’t know it,” says MIT researcher Keith Mohr.

Educators encourage their students to question the authority of algorithms.

In one of Ms. Schumann’s classes, female students were asked to experimentally determine how popular algorithms, often developed by white or Asian men, had difficulty recognizing African American faces. This dysfunction led to the arrest of many innocent black residents of the Bronx.

In Ms. Schumann’s experiment, Amazon’s facial recognition algorithm determined with 76.5% certainty that Oprah Winfrey was male. Microsoft’s equivalent system decided that Michelle Obama was “a young African-American man in a black top,” while IBM’s system informed users with an 89% chance that the photo of tennis player Serena Williams was a man.

The goal of the course is to convey to children that computer algorithms can be flawed, like cars and any other human construct, by encouraging young people to ask questions about problematic technological tools.

Simple Presentation

The course, conceived and developed by chatbot ChatGPT on the topic of wearable technology such as digitally connected smartwatches, was impressively detailed and thorough, with introductory discussion, reading of wearable technology sources, exercises, and a conclusion. Ms. Schumann asked her students to follow the ChatGPT program for 20 minutes as if it were a real lesson.

At the end of the lesson, the girls groaned in boredom and commented that the algorithm lesson was simplistic, repetitive and reminiscent of an advertising company presentation. The class unanimously agreed that Ms. Schumann’s lesson was better than the chatbot lesson. “Do you think your teachers should use ChatGPT?” Mrs Schuman asked. All the students responded with a resounding “no”.