(Quanta Magazine) – By injecting “thoughts” directly into large language models, researchers are trying to map out AI’s capacity for introspection.

Now, after over a century of studying introspection in humans and animals, researchers are beginning to explore the phenomenon in unlikely new subjects — large language models. These systems, which are based on mathematical structures called neural networks, excel at cognitive tasks that until recently seemed exclusive to humans. Can they introspect as well?

In 2025, Jack Lindsey, a researcher at Anthropic, devised a way to test introspection in language models by injecting “thoughts” into their neural networks — in a kind of targeted digital neurosurgery. Lindsey’s experiment demonstrated that models could at least sometimes detect and describe these injections. It looked a lot like introspection. (Read More)