AI-text detection tools are really easy to fool The news: As soon as ChatGPT launched, there were fears that students would use the chatbot to churn out passable essays. In response, startups started creating products that promise to spot whether text was written by a human or a machine. The problem is, it's relatively simple to trick these tools and avoid detection, according to new research. How it unfolded: Researchers tested 14 commonly-used AI-detection tools, and found that while they detected human-written text consistently, they struggled to pick up ChatGPT-generated text that had been slightly rearranged by humans or obfuscated by a paraphrasing tool. This suggests that all students need to do is slightly adapt the essays the AI generates to evade detection. Why it matters: Experts believe that overreliance on AI detection tools could lead to incorrect accusations of cheating, which could have dire consequences for students' academic career. Read the full story. —Rhiannon Williams Read more about how AI will change education in this great piece by my colleague Will. |
Комментариев нет:
Отправить комментарий
Примечание. Отправлять комментарии могут только участники этого блога.