AlgoTrading

AChatGPT, a powerful AI software capable of writing sophisticated responses to prompts, arrived on the scene last year and has prompted intense speculation about the long-term repercussions on a host of industries and activities. Schools and universities have been forced into a fundamental rethink of how they conduct tuition and academic testing. Concerns about AI-based plagiarism have pushed a number of institutions to opt for an outright ban of bots like ChatGPT, but enforcing this is difficult. Some educators are instead experimenting with ways to use generative AI to enhance their lessons. This raises the question of whether it is possible to open the door to AI in education without undercutting the most important features of human learning. Accuracy is one of the major challenges with generative AI, as language models are known to “hallucinate”, which means they fabricate facts, sources and citations in unpredictable ways. There is also evidence of bias in AI-written text, including sexism, racism and political partisanship. The "black box" effect, which means it is not clear how AI comes to its conclusions, is a major concern for students. At the Virginia Military Institute, a history lecturer found that AI can give false information. At the Institut auf dem Rosenberg, one of Switzerland's most elite boarding schools, 12th-grade student Karolina was working on an assignment for her sociolinguistics class and handed in a video featuring an analytical dialogue between two women in the relevant accents. The script was based on research and was generated by Colossyan Creator, AI software from a London-based start-up. One of researches encouraged students' use of AI tools, following other qualification bodies such as the International Baccalaureate and Wharton, the University of Pennsylvania's business school. "Children are using tech to study and learn, with or without AI," she says.