Enlarge / AI generated image of a child using a computer.
Ars Technica
New York City public schools have blocked access to OpenAI’s ChatGPT AI model on their network and devices, education news site Chalkbeat reports. The move comes amid fears from educators that students will use ChatGPT to cheat on assignments, inadvertently introduce inaccuracies into their work, or write essays in a way that distracts them from learning the material.
ChatGPT is a large language model developed by OpenAI and is currently accessible through any web browser for free during testing. Humans can use it to write essays, poetry, and technical documents (or even simulate a Linux console) at a level that can often pass for human writing — although it can also produce a very confident-sounding but inaccurate Results.
Per Chalkbeat, NYC Department of Education spokeswoman Jenna Lyle said, “Due to concerns about negative impacts on student learning and concerns about the security and accuracy of content, access to ChatGPT on New York City’s networks and devices is public Schools Limited While the tool can provide quick and easy answers to questions, it does not build critical thinking and problem-solving skills that are essential for academic and lifelong success.”
advertising
In September, pedagogical concerns about large language models surfaced after someone on the OpenAI subreddit claimed they were using GPT-3 (the underlying technology behind ChatGPT) to write essays and answer questions for school projects. At that time, GPT-3 was only accessible via an API or a dedicated user interface through the OpenAI website and for a usage-based fee. After OpenAI unveiled ChatGPT to a much larger audience for free on November 30th, calls of alarm spread among educators.
As a language prediction model, ChatGPT is a type of neural network trained on the text of millions of books and websites. The model’s “knowledge” of typical text constructions helps it predict the most likely output after someone enters a prompt. So given a prompt like “Mary had a,” it could complete the sentence with “Little Lamb,” pulling from the top statistical associations between words the AI model encountered during the training process. During a session with ChatGPT, all of the conversation history is the prompt that the model tries to complete, albeit in conversational form.
Although ChatGPT is blocked on computer networks in New York public schools, Chalkbeat reports that individual schools can request access to ChatGPT to study the technology behind it. And while it’s unclear if ChatGPT will have the feared impact on education, this move by the largest school district in the US could inspire others to enact similar bans on the technology.