Why does ChatGPT scare teachers?
The development of AI systems has brought new challenges to their use. The most prominent of these systems is the interactive chatbot (ChatGPT), of which OpenAI launched a beta version a few weeks ago. Teachers and professors at universities and research institutes allow students to cheat on a large scale without being detected.
Unlike all the other interactive chatbots available, ChatGPT chats with users in a very flexible way, providing human-like answers to difficult questions as if they were talking to another person, and is prone to inappropriately refusing to answer questions like how to bully and how to burgle a house.
ChatGPT Bots and Classroom Cheating:
The buzz generated by the ChatGPT bot prompted teachers to start trying out the bot and see what they encountered. It was very close to what you would expect from a file.
Educators have wondered if students could use ChatGPT to create reports, essays, and programming exercises without their knowledge. How would the proliferation of such tools affect their academic performance?
Teachers usually give students the task of writing an essay. This is because it teaches students many important skills such as researching essay topics, classifying and judging arguments, and collecting the information they find and expressing it in their own words.
“It’s a very, ...
ChatGPT beta reached 1 million users in just 5 days. Some users asked the usual questions like: for example, how do you explain to a 6-year-old that Santa Claus does not exist, or how do you ask for new ideas for birthday parties. Some asked the bots to find problems in the code they wrote, and others asked them to write poetry and essays on various topics.
So, for some students, the ChatGPT bot has become a magical tool for doing their homework, as one student told a newspaper, “I used ChatGPT twice to do my homework.” The idea came to him after he saw people on Twitter explaining how powerful the tool was after its release on November 30, 2022.
“His homework was in a computer science language he didn’t know. When he asked ChatGPT about it, he got an answer almost immediately. He gave it to the teacher.
Later that day, he used ChatGPT to help write a homework question for his teacher. The students also said they planned to use ChatGPT to cheat on exams instead of Chegg, the homework help site they had used in the past.
Students also stated that they were not concerned about being discovered because they did not believe that teachers could discover their answers and determine that they were the result of a computer program. As such, teachers face an ethical dilemma in the use of AI and must deal with it.
Artificial Intelligence and Moral Hazard:
OpenAI developed the ChatGPT chatbot using a large-scale language model (GPT-3.5) developed over many years to predict language, and it was trained to predict the next word in a sentence by finding patterns, often distinguishable from human text.
The initial development of the ChatGPT bot required an AI trainer to deliver a conversation that would act as both a user and an AI assistant, often humorous or philosophical.
However, some of its answers were completely wrong or biased. Content writers acknowledge that ChatGPT bots are not perfect and sometimes give misleading answers.
Teachers assume that ChatGPT bots and similar AI tools will improve over time, increasing students’ knowledge. Some say teachers are adjusting assessments to account for the potential for cheating. For example, some say they ask students to do their homework during class, while others might figure out how to write a question that requires more thought. For bots, the problem is more difficult.
Many teachers only learn to write when students wrestle with ideas and put them into sentences, because students start to get lost in what they mean and figure it out as they write, and I agree that you can.
“The process of writing changes our cognition,” says Joshua Wilson, a professor of education at the University of Delaware.
David K. Thompson, Associate Professor of History at Sacred Heart University in Connecticut, USA. “From my experience with the bot (ChatGPT), I found it to work well with the types of questions that come up in many homework assignments. The question to ask: Economic revolution, but how did Frederick Douglass present his arguments against slavery when asked the more complex question “No,” he said. Professors need to present not only verifiable facts, but also problems that govern analytical thinking.
At the same time, Joshua Eller, an assistant professor at the University of Mississippi, said: “In teaching mathematics, this is a moment like changing the way writing is taught.”
As teachers figure out how to live with new AI tools, some companies are thinking about how to get around them.
Turnitin, which develops software widely used by students and researchers to detect non-original content in academic papers to prevent plagiarism, is currently looking at ways to detect material generated by artificial intelligence. I’m here.
“Automated essays are different in many ways from student-written work because students write in their own way, which is not in ChatGPT content,” a company official said.
AI company Hugging Face, Sasha Lucioni, a research scientist at
OpenAI itself is looking for a solution.
According to OpenAI researcher Scott Aronson, the company is exploring different ways to combat abuse, including using watermarks and providing tools to detect ChatGPT answers.
Some users have questioned the usefulness of the watermark solution and whether it will be sufficient (Aaronson).
ChatGPT had its own idea for a solution, but when asked how to combat potential cheating, the bot offered a few suggestions, including: Give students the support they need so they don’t have to cheat.
Finally, Pott explained that it is important for teachers to clearly communicate their expectations for academic integrity to their students and to take steps to prevent misconduct. This step helps build a culture of honesty and integrity in the classroom.
Despite the ethical dilemmas, why do we use AI?
The technological revolution is changing our lives at a breakneck speed, dramatically changing the way we work, learn and live together. There is no doubt that artificial intelligence is at a huge stage of development. During this current period. Many new applications and tools have emerged in many fields, including security, environment, research, education, health, arts and commerce.
Artificial intelligence is expected to change many of the jobs that humans will soon do, so the right approach to these technologies is to learn how to use them for the betterment of humanity.
Since AI will lead to new forms of human civilization, we need to ensure that the AI revolution is developed through a humanistic, value-based approach, as it will open up new horizons for human civilization, but we must ensure that its development has ethical dilemmas that require careful consideration.