A New AI Era in Education
In the hallowed halls of academia, a new player has entered the game. ChatGPT, a creation of OpenAI, is making waves, and not just the ripples you’d expect from a pebble in a pond. We’re talking tsunamis. A recent study by the University of California, Los Angeles (UCLA) has shown that this machine’s problem-solving capabilities are on par with, if not exceeding, those of undergraduate students.
The Experiment: Man vs Machine
The psychologists at UCLA decided to put GPT-3 to the test, pitting it against complex arrays of shades and predicting the next image. The result? A performance that rivals the brightest minds in American colleges, solving 80 percent of reasoning problems. The undergraduates? They lagged behind at a mere 60 percent.
The Implications: A Brave New… Wait, Scratch That
Published in the Nature Human Behaviour journal, this research has sent shockwaves through the education sector. SAT exams, the gateway to US college admission, were conquered by this machine. But before we herald the end of human intelligence, let’s take a step back.
The model stumbled when asked to connect passages of prose with different short stories conveying the same meaning. Here, the students outperformed the machine. Social interactions, mathematical logic, and problem-solving skills like moving sweets from one bowl to another proved to be stumbling blocks.
“It’s definitely not fully general human-level intelligence. But it has definitely made progress in a particular area,” said Taylor Webb, a postdoctoral researcher in psychology at UCLA.
The Concerns: Cheating and Critical Thinking
The emergence of ChatGPT has raised eyebrows and concerns. Universities are restricting access, fearing student plagiarism. A survey revealed that 50 percent of students in the US use AI tools for college assignments, with 17 percent not even bothering to edit after using AI.
But let’s not jump to conclusions. Experts argue that AI will not affect critical thinking and problem-solving skills. Jenna Lyle, a spokeswoman for the New York City Department of Education, emphasized that ChatGPT fails to build critical thinking and problem-solving abilities, crucial for academic achievements.
The Future: Embracing the Change
Last month, Harvard announced a groundbreaking move: computer science students will use AI to find errors in their codes. The CS50 course will be taught entirely by a chatbot. The future is here, and it’s time to embrace it.
