Top News

Suspense crime, Digital Desk : Generative AI tools like ChatGPT have become the ultimate productivity hack, helping us draft emails, write code, and summarize complex documents in seconds. But is this convenience coming at a hidden cost to our own cognitive abilities? A fascinating new study from researchers at MIT suggests the answer is yes, revealing how over-reliance on AI can make us less sharp.Researchers at MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) set up an experiment to test how AI assistance affects human performance. They asked human proofreaders to check texts for errors. One group worked alone, while the other used an AI assistant designed to help them.Here’s the catch: the researchers deliberately programmed the AI to sometimes miss certain types of errors. The results were startling.While the group using the AI assistant worked faster, they were significantly less accurate. More importantly, they fell into a trap known as “automation complacency”—they began to trust the AI’s output more than their own judgment. Over time, the AI-assisted workers actually became worse at spotting the very errors the AI was programmed to miss. Their own proofreading skills started to degrade because they were outsourcing the thinking to the machine.This phenomenon is called “cognitive offloading.” Just as relying on GPS can weaken our natural ability to navigate, handing over mental tasks like writing and editing to AI can cause our own skills in those areas to atrophy from disuse. The study highlights that when humans and AI work together, the human often ends up simply “supervising” the AI, which can lead to a dangerous level of inattention and skill decay.The findings don’t suggest we should abandon these powerful tools. Instead, they serve as a critical reminder of the importance of mindful engagement. The study’s authors warn that as AI becomes more integrated into high-stakes professions like medicine and finance, the risk of human skill erosion is a serious concern.The key takeaway is that AI should be treated as a co-pilot, not an autopilot. While it can handle the heavy lifting, the final judgment, critical thinking, and responsibility must remain with the human user to avoid falling into the automation trap.