Subliminal Learning

"Student" Als pick up unexpected traits such as a love of owls from their "teachers"

- Emma R. Hasson

FROM A TEACHER'S body language, inflection, and other context clues, students often infer subtle information far beyond the lesson plan. And it turns out artificial-intelligence systems can do the same—apparently without needing any extra context. Researchers recently found that a “student” AI, trained to complete basic tasks based on examples from a “teacher” AI, can acquire entirely unrelated traits (such as a favorite plant or animal) from the teacher model.

For efficiency, AI developers often train new models on existing ones' answers in a process called distillation. Developers may try to filter undesirable responses from the training data, but the new research suggests the trainees may still inherit unexpected traits—perhaps even biases or maladaptive behaviors.