Riley essay: AI is a 'cognitive hot dog' for the brain

In an essay for The Verge, education writer Benjamin Riley calls AI "a cognitive hot dog." "The occasional hot dog won't harm anyone," he writes, "but it sure will if it becomes a regular meal."

Riley, who runs a think-and-do tank called Cognitive Resonance, argues the AI industry keeps misjudging the damage because it treats brains as computers. He points to Google DeepMind's Demis Hassabis calling the brain "a biological approximation to a Turing machine," and to Elon Musk telling people to "think of the brain as a biological computer." The alternative Riley reaches for is Paul Cisek, a University of Montreal neuroscientist, who models brains as feedback-control systems that act to shape what they sense, grown out of evolutionary history rather than input-output.

The evidence Riley lines up is domain-specific. A working paper tracking 26,811 Chinese secondary pupils over thirty months found homework scores rose while monthly exam scores fell within six months of AI adoption. An ALEKS mathematics analysis covering 3.2 million interactions showed college students' study time on AI-susceptible problems fell 26.9% over eleven quarters, with no change when they were proctored. An MIT Media Lab study of 54 students found ChatGPT users showed weaker engagement and recall when later asked to write unaided. The researchers called it "cognitive debt."

Policy has started to follow. Norway in June told schools that children aged 6 to 13 should, as a rule, not use AI. New York City and Los Angeles, the two largest US school districts, paused most student AI use for this school year. Riley does flag a counter-case: a Harvard randomized trial, he writes, found custom-built AI physics tutors produced more than twice the learning of excellent active-learning classes. Two researchers we track shared the piece the day it ran.