The artificial intelligence chatbot has surprised users / OpenAI. With the invention of the camera, artists could create images without lear...
 |
| The artificial intelligence chatbot has surprised users / OpenAI. |
ChatGPT is extraordinary, but its responses are algorithmic. Already, plagiarism-detection services are adding features to detect AI-generated text. Educators may closely scrutinize students’ submitted work for signs of AI support, or conversely might embrace AI as a tool to assist students’ writing. But ultimately, ChatGPT won’t supplant educators’ focus on cultivating the writing abilities of their students.
Nor should ChatGPT supplant this focus. Even if the program’s responses were truly indistinguishable from a student’s, there is value in learning how to write. Individuals should trust their own ideas, not those collected and generate by a computer. Bold ideas are bold precisely because they are unconventional. They run counter to society’s accepted knowledge. Perhaps ChatGPT will have its impact on education by motivating educators to emphasize to their students the importance of self-determination.
The New Google
The release of ChatGPT came at a serendipitous time, right when college students were studying for final exams or turning in final essays. I have seen the AI write love poems, give a detailed summary of an excerpt, write full sets of code, and even draw up a nondisclosure agreement. These new tools might become the new Google.
If the databases are constantly being updated with current news and information, as well as connected to the internet, we could use AI to learn and solve problems in daily life. When I went to look up an advanced organometallic chemistry topic, ChatGPT gave a better summary than Google. College professors will have to determine how they want to proceed and if they need to have in-person final essays without technology. But without technology in the classroom, will teaching regress?
AI tools such as ChatGPT can help users achieve specific goals. There is always concern about new technology and the resulting potential paradigmatic shifts. But history will remind us that it’s important to acknowledge these technological developments and educate about the strengths and weaknesses of these tools. It’s equally important, however, not to forget the basics.
ChatGPT can’t replace reasoning or critical thinking. While AI tools can make essays read better, they can’t replace knowing how to form thoughts into careful arguments. The most significant challenge for future educators is finding out how best to develop and assess those skills.
New York City public schools banned access to ChatGPT, an artificial-intelligence chatbot, on its internet networks and school devices after officials raised concerns that students could use the AI program to answer questions, do homework or write essays.
ChatGPT quickly drew attention from the public and students after its November release, with some industry observers calling it one of the most intelligent AI applications ever created. The program crossed a million users a few days after its launch. And its popularity has been a boon to its developer, OpenAI, which the Journal reported this week is in talks to sell shares at a $29 billion valuation. That would make it one of the most valuable U.S. startups, on paper.
ChatGPT could upend entire industries and schools by automating certain jobs or offering intelligent answers to almost any question. Many students have delighted in it, while teachers have panicked. The chatbot’s answers are often so colloquial that it can be difficult for teachers to know if a student has used the program to cheat. But the chatbot doesn’t always provide accurate information.
Medieval Lessons
Live cameras, screen recordings and antiplagiarism software are all too familiar to the current university student. As technology advances, such defenses will continue to be deployed against the illicit use of new tech in the academy. An unceasing tit-for-tat will ensue between tools such as ChatGPT and security measures to curtail academic dishonesty. Educators may strive to stay ahead of all such obstacles, but this is a losing battle.
There is another way: Study with Catholic friars. The friars follow the format of a scholastic studium, an educational model that uses formalized arguments as the primary method of teaching. Many exams are given orally, a mode that requires clear thinking and concise speaking on the part of the student. Papers are not submitted but presented to the class. Theses are defended while friars hurl objections and counterpoints at the student. In such rhetorical exercises, there is no opportunity to hide behind clever AI. Moderns can learn much from medieval ways.
An Auxiliary Resource
The ChatGPT bot can be used for the benefit of the students, or it can be used to their detriment. The outcome will depend on how well faculty can integrate this technology into their curricula, as well as the integrity of the students to use it properly. The obvious concern is academic fraud. Educators will need to implement new assessment methods to mitigate cheating.
Written in-class assignments might become more common. Instead, students should use AI tools as auxiliary resources. Even if conversational AI is only semi-reliable at this point, it can be used to learn about new topics, or ask questions outside class. The adjustment period will come as a shock to the education system. This is normal for major changes throughout history, such as the Gutenberg Press, the internet or the personal computer. We can remain optimistic, however, that the good faith of most students and faculty will make this technological advancement a net positive.
New York City’s Department of Education, which runs the largest school district in the country, said this week that it had “concerns about negative impacts on student learning and concerns regarding the safety and accuracy of content.” “While the tool may be able to provide quick and easy answers to questions, it does not build critical-thinking and problem-solving skills, which are essential for academic and lifelong success,” said Jenna Lyle, a spokeswoman for the department.
An OpenAI spokeswoman said the company was developing measures to help people identify text generated by ChatGPT. “We don’t want ChatGPT to be used for misleading purposes in schools or anywhere else,” she said. ChatGPT has raised concerns in schools. Students could ask the chatbot to write code or craft prose about any topic, like equating the movie “Ferris Bueller’s Day Off” to an existentialist text. New York City’s education department appears to be the only one in the U.S. to restrict access to the program.
Educators have said that while ChatGPT could help students cheat, it could also be used as a search engine to help them learn new information. ChatGPT has been banned from other places. Stack Overflow, an online community for software developers, has temporarily banned users from posting any text or code generated by ChatGPT. One of the world’s largest machine learning conferences, the International Conference on Machine Learning, has banned papers written using the program.
The restrictions may be hard to enforce in New York City schools. Education officials can’t stop students—or teachers—from opening the chatbot on Wi-Fi networks or devices that aren’t affiliated with their schools. They could also potentially use their cellular network on campus to access ChatGPT.
New York City’s education department said this week that individual schools could request that the department grant access to ChatGPT. Chalkbeat, an education news site, earlier reported the ban. OpenAI released ChatGPT at a challenging time for U.S. public education.
U.S. public schools have lost more than a million students since the pandemic began, forcing some districts to close underused schools. School officials blamed the enrollment decline on falling birthrates, a rise in homeschooling and growing competition from private and charter schools.