ChatGPT creates a new tool to catch students cheating using ChatGPT

ChatGPT creator OpenAI’s logo is seen on a mobile phone in New York on Tuesday, Jan. 31, 2023. Richard Drew-AP Images

The maker of ChatGPT is trying to curb its reputation with a new tool that lets the teacher know whether a student or artificial intelligence wrote that homework.

The new AI Text Classifier launched by OpenAI on Tuesday is the following. Weeks of discussion in schools And colleges chatGPT out of concern that the ability to write about anything on command could undermine academic integrity and hinder learning.

It warns that OpenAI A new device – Like others already have – it’s not stupid. AI-powered text recognition is “imperfect and sometimes wrong,” said Jan Lake, head of the OpenAI Alignment Team, which aims to make systems safer.

“Because of that, it shouldn’t be relied upon when making decisions,” Lake said.

Teenagers and college students were among the millions of people who started experimenting with ChatGPTT after it launched as a free app on OpenAI’s website on November 30. And while many have found ways to use it in creative and harmless ways, the fact that it can easily answer take-home test questions and help with other assignments has sparked panic among some teachers.

As schools reopened for the new year, New York City, Los Angeles and other large public school districts began banning the use of devices in classrooms and at school.

The Seattle Public Schools District initially blocked ChatGPT from all school devices in December, but has opened access to teachers who want to use it as an instructional tool, said district spokesman Tim Robinson.

“We can’t afford to ignore it,” Robinson said.

The district is discussing possibly expanding the use of ChatGPT into the classroom to help teachers train students to be better critical thinkers and to have students use the application as a “personal tutor” or to help generate new ideas while working on an assignment. Robinson said.

School districts around the country say they are rapidly evolving the conversation around ChatGPT.

“The first reaction was, ‘OMG, how are we going to get rid of all the waves of cheating on ChatGPT,'” says Devin Page, technology specialist for the Calvert County Public School District in Maryland. Now the awareness that “this is it” is growing and banning it is not the solution, he said.

“I think we’d be naive if we didn’t recognize the dangers of this tool, but if we could prevent our students and us from using it for all its potential,” said Page, who is district-minded. Like itself, ChatGPT will eventually get unblocked, especially if the company’s detection service is activated.

OpenAI emphasized the limitations of its detection tool in a blog post on Tuesday, but said that in addition to preventing theft, it can help do so. Discover automated disinformation campaigns And abusing AI to impersonate other people.

The longer the text, the better the device will be able to detect whether an AI or a human wrote something. Type in any text — a college admissions essay, or a literary analysis of Ralph Ellison’s “The Invisible Man” — and the tool will label it as “highly unlikely, improbable, vague, possible, or probable” generated by AI.

But like ChatGPT itself, He was trained In digitized books, newspapers, and online articles, he often spouts lies or nonsense with confidence, and it is not easy to interpret how he got there.

“We don’t know what kind of design it’s going to focus on or how it’s going to work internally,” Lake said. “We don’t have much to say at this point about how the classifier actually works.”

Higher education institutions around the world are also beginning to discuss the responsible use of AI technology. SciencePo, one of France’s most prestigious universities, banned its use last week and warned that anyone caught using ChatGPT and other AI tools to secretly obtain written or oral work could be banned from SciencePo and other institutions.

In response, OpenAI said it has been working for several weeks to develop new guidelines to help teachers.

“As with many other technologies, a district may decide that it’s not appropriate for use in their classroom,” said Lama Ahmed, an OPNI policy researcher. “We certainly won’t beat them one way or the other. We want to give them just the information they need to make the right decision.

It’s now a rare public role for a research-oriented San Francisco startup. Backed by billions of dollars in investment Interest from partner Microsoft and from the public and governments is growing.

French Digital Economy Minister Jean-Noel Barot recently met with OpenAI executives in California, along with CEO Sam Altman, and told an audience at the World Economic Forum in Davos, Switzerland, a week later, that he was optimistic about the technology. But the government minister – a former professor at the Massachusetts Institute of Technology and the French business school HEC in Paris – also has difficult ethical questions.

“So if you’re in law school, you have room for concern because obviously ChatGPT can provide some pretty impressive challenges compared to other tools,” he said. “If you’re in an economics faculty, you’re fine because chatgpty is hard to meet or deliver when you’re in a graduate-level economics faculty.”

He said it will be even more important for users to understand the basics of how these systems work so they know what biases might exist.